diff --git a/docs/reference-matrix.md b/docs/reference-matrix.md index 4be14b07c..bc099b38e 100644 --- a/docs/reference-matrix.md +++ b/docs/reference-matrix.md @@ -248,6 +248,7 @@ SSD/OS systems columns. | JiusiServe/vllm-evolve @ 43cc82c | search | scheduling-algorithm evolution, search policy | `optimization-search-policy`, `dense-performance-evidence`, `inference-benchmark-design` | | [OpenBMB/JustGoal repo + paired paper](https://github.com/OpenBMB/JustGoal/tree/f052fc0b28fc7f10b2ef1ae7aa4682aabbbe868d) @ `f052fc0b28fc7f10b2ef1ae7aa4682aabbbe868d` (repository Apache-2.0 default; NOTICE BSD exceptions; paper license UNKNOWN) | evaluation | Bounded direct inspection, 2026-11-02: complete 21-page paper/appendices and selected DSA wrapper/CUDA workspace/stream contracts; [exact readset and exclusions](../src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/justgoal-source.json). Source-only methodology; author-reported comparisons, released failures versus repaired timing and official/stricter oracle differences retained | [inference-benchmark-design reference](../src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/justgoal-evaluation-contracts.md); candidate/not_run, efficacy not_evaluated; no new implementation/backend coverage or full campaign runner | | [AutoTuneBench repo + paper v1](https://github.com/li-ch/autotunebench/tree/64cc8d7ca5d0ee40c4e2189b2c2c8dfb1a2a7c2f) @ `64cc8d7ca5d0ee40c4e2189b2c2c8dfb1a2a7c2f`; [arXiv:2609.18123v1](https://arxiv.org/abs/2609.18123v1) (selected repository code/data MIT; paper CC BY 4.0) | evaluation | Bounded source inspection, 2026-11-02: treatment delivery, arm-validity precedence, exported protocol/check/record gaps, weak-denominator and anomaly counterexamples; complete 15-page paper, 6 figures, 7 tables and [exact selected readset/rights](../src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/autotunebench-source.json) | [inference-benchmark-design reference](../src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/autotunebench-trial-validity.md); source-derived candidate/not_run, efficacy not_evaluated; invalid-arm/source-median conflicts retained, no new controller or measured implementation coverage | +| [Neroued/ninfer](https://github.com/Neroued/ninfer/tree/1882c6adbc98043fcf89f3589aa5addb3742777a) @ `1882c6adbc98043fcf89f3589aa5addb3742777a` (selected software/docs Apache-2.0) | evaluation | Bounded source inspection, 2026-11-03: Qwen3.5 GDN/MTP eager record/fold and checkpoint/adoption/retirement contracts, unexecuted fold-test definitions and historical completion denominators; [27-file readset and rights](../src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/ninfer-source.json) | [`inference-benchmark-design` reference](../src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/ninfer-state-and-completion-contracts.md); candidate/not_run, efficacy not_evaluated. Historical author results retain their separate execution revision; no new implementation, backend or measured resource coverage | | Dong1017/OmniRSI @ `codex/skill-eval-bootstrap` (82218b4) | evaluation | skill_distill / skill_eval harness, no_skill/human/distilled contract, Tier A pilot | `npu-skill-factory` (evaluation contract) | | InferMatrixLab/InferRSI PR #9 @ 4ad551a9 | reference | `skill_library/` layout + knowledge records | `npu-skill-factory` (bundle layout) | | RRSI — arXiv 2609.24972v3 | search; mined via reviewed Pack F | bounded harness proposals; calibrated noise/cost/novelty gates; grouped-ablation limits, replacing the earlier generic constraint summary | `optimization-search-policy`; [exact ledger](../src/inferrsi/data/skill_library/methods/optimization-search-policy/references/papers-20261015/references/source-notes.md#f1--rrsi) | diff --git a/docs/searchspace-matrix.md b/docs/searchspace-matrix.md index aad404af3..60d59962a 100644 --- a/docs/searchspace-matrix.md +++ b/docs/searchspace-matrix.md @@ -600,6 +600,12 @@ NOASSERTION and InferRSI's repository license. | --- | --- | --- | --- | | Accelerator (methodology only) | Verify: trace treatment to the eligible consumer, apply frozen arm validity before interpreting a null, and separate declared protocol, reached checks and persisted gate evidence | [inference-benchmark-design reference](../src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/autotunebench-trial-validity.md) and [source record](../src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/autotunebench-source.json); AutoTuneBench repo + paired paper | Source-derived candidate / empirical not_run; paper measurements remain author-reported, efficacy not_evaluated. Contextual source bars are not InferRSI defaults. No new implementation, measured hardware/resource coverage, backend compatibility, or RAM/Storage/Network/OS coverage. | +## NInfer state and completion validation (source-derived) + +| Resource context | Stage / scoped question | Canonical owner | Evidence boundary | +| --- | --- | --- | --- | +| Accelerator / Host / service boundary (methodology only) | Observe / Verify: freeze coupled checkpoint and same-block state oracle; distinguish zero accepted drafts from zero adopted inputs, logical commit from device retirement, and selected steady intervals from complete-cohort/termination accounting | [inference-benchmark-design reference](../src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/ninfer-state-and-completion-contracts.md) and [source record](../src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/ninfer-source.json); NInfer `1882c6adbc98043fcf89f3589aa5addb3742777a` | Source-derived candidate / empirical not_run; unexecuted test definitions and historical author results with a separate execution revision. No new implementation, measured hardware/resource coverage, backend compatibility, native state/model/quality/performance result or service acceptance; efficacy not_evaluated. | + ## TorchNPU framework source procedures | Resource | Stage / mechanism | Canonical owner | Evidence boundary | diff --git a/docs/skill-coverage.md b/docs/skill-coverage.md index 50c3f1b6c..2a6ceee0b 100644 --- a/docs/skill-coverage.md +++ b/docs/skill-coverage.md @@ -4,7 +4,7 @@ Generated from the registry, the skill sidecars and the two maintained matrices. **Do not edit by hand**; regenerate after any registry, skill, matrix or frozen-plan change. -- Input digest: `sha256:7640b171256091efd1514a33d5e434eef3feaed238fc92fe3eeba02fee94e9ff` +- Input digest: `sha256:ba1e64967e2b3a1eef429bca60ff21c3a78d3be4cdbed077ab255dbc40643cee` - Regenerate: `python tools/build_skill_coverage.py` - Verify without writing: `python tools/build_skill_coverage.py --check` @@ -131,7 +131,7 @@ optimization is fully covered or that any procedure has measured benefit. | [choose-parallel-training-plan](../src/inferrsi/data/skill_library/methods/choose-parallel-training-plan/SKILL.md) | common | Compare stage, device-mesh and operator-sharding candidates for static distributed training while accounting for ... | Comparing stage, device-mesh and operator-sharding candidates for static distributed ... | failure_modes, input_contract, produced_artifact, refusal_point
3 advice | 1 Accelerator · Diagnose / Configure / Optimize / Verify
4 Network · Diagnose / Configure / Optimize / Verify | 1 · Systems sources (Alpa — [2201.12023v3](https://arxiv.org/abs/2201.12023v3)) | candidate · not_run · 1 readset(s) | no-eval | | [dense-performance-evidence](../src/inferrsi/data/skill_library/methods/dense-performance-evidence/SKILL.md) | common | Taxonomy of performance evidence an optimization agent needs, ordered by collection cost and diagnostic power. Maps ... | Deciding which performance evidence to collect next and which claim each evidence level ... | failure_modes, input_contract, produced_artifact, refusal_point
4 advice | 1 Accelerator · Observe / Profile / Verify
2 RAM/Host · Verify
4 Network · Verify | 1 · Systems sources (NVIDIA/TensorRT-LLM @ f33e0e5fd)
2 · Methodology and corpus references (JiusiServe/vllm-evolve @ 43cc82c)
2 · Methodology and corpus references (microsoft/SkillOpt @ 343db22) | candidate · not_run · 0 readset(s) | — | | [evaluate-elastic-serving-placement](../src/inferrsi/data/skill_library/methods/evaluate-elastic-serving-placement/SKILL.md) | common | Compare migrating an LLM request between model replicas with changing its sequence-parallel execution group, including ... | Comparing migration of a running LLM request between compatible model replicas with ... | failure_modes, input_contract, produced_artifact, refusal_point
3 advice | 1 Accelerator · Diagnose / Configure / Optimize / Verify
2 RAM/Host · Diagnose / Configure / Optimize / Verify
4 Network · Diagnose / Configure / Optimize / Verify | 1 · Systems sources (Llumnix — [2406.03243v1](https://arxiv.org/abs/2406.03243v1)) | candidate · not_run · 1 readset(s) | no-eval | -| [inference-benchmark-design](../src/inferrsi/data/skill_library/methods/inference-benchmark-design/SKILL.md) | common | Select task quality gates, reference parity checks and performance workloads for text, agentic coding, work agents ... | Selecting task quality gates, reference parity checks and performance workloads for text ... | failure_modes, input_contract, produced_artifact, refusal_point
5 advice | N/A: SKILL.md §§ Select the task and its evaluation contract, Statistical requirements, Shared-machine measurement discipline, and Search integrity and final-state verification design comparable evaluation; hardware identity and allocation isolation are comparison controls, not a hardware-resource mechanism. | 2 · Methodology and corpus references (JiusiServe/vllm-evolve @ 43cc82c)
Task-specific benchmark selection (issue #60) ([Issue #60 report](https://github.com/InferMatrixLab/InferRSI/issues/6 ...)
Task-specific benchmark selection (issue #60) ([NVIDIA/NeMo-Speech.cpp @ 20999cc9](https://github.com/NVIDIA/NeMo-Spe ...) | candidate · not_run · 7 readset(s) | — | +| [inference-benchmark-design](../src/inferrsi/data/skill_library/methods/inference-benchmark-design/SKILL.md) | common | Select task quality gates, reference parity checks and performance workloads for text, agentic coding, work agents ... | Selecting task quality gates, reference parity checks and performance workloads for text ... | failure_modes, input_contract, produced_artifact, refusal_point
6 advice | N/A: SKILL.md §§ Select the task and its evaluation contract, Statistical requirements, Shared-machine measurement discipline, and Search integrity and final-state verification design comparable evaluation; hardware identity and allocation isolation are comparison controls, not a hardware-resource mechanism. | 2 · Methodology and corpus references (JiusiServe/vllm-evolve @ 43cc82c)
2 · Methodology and corpus references ([Neroued/ninfer](https://github.com/Neroued/ninfer/tree/1882c6adbc9804 ...)
Task-specific benchmark selection (issue #60) ([Issue #60 report](https://github.com/InferMatrixLab/InferRSI/issues/6 ...)
Task-specific benchmark selection (issue #60) ([NVIDIA/NeMo-Speech.cpp @ 20999cc9](https://github.com/NVIDIA/NeMo-Spe ...) | candidate · not_run · 9 readset(s) | — | | [inference-optimization-casebook](../src/inferrsi/data/skill_library/methods/inference-optimization-casebook/SKILL.md) | common | How to structure and use a casebook of past optimization decisions as a decision-support tool. Extracts the casebook ... | Structuring and querying a casebook of past optimization decisions as decision support ... | failure_modes, input_contract, produced_artifact, refusal_point
4 advice | 1 Accelerator · Optimize | 1 · Systems sources (NVIDIA/TensorRT-LLM @ f33e0e5fd) | candidate · not_run · 0 readset(s) | no-eval | | [kernel-writing-guide](../src/inferrsi/data/skill_library/methods/kernel-writing-guide/SKILL.md) | common | Decision guide for writing custom inference kernels: when to write one, which framework to choose (Triton, CuTe DSL ... | Deciding whether to write a custom inference kernel, routing a kernel pattern to Triton ... | failure_modes, input_contract, produced_artifact, refusal_point
6 advice | 1 Accelerator · Optimize / Diagnose / Configure / Verify | 1 · Systems sources (Dao-AILab/causal-conv1d @ `cd81f0413cad2fc1e6f17e785ac39f59aae690cd` ...)
1 · Systems sources (NVIDIA/TensorRT-LLM @ f33e0e5fd)
1 · Systems sources (huggingface/transformers @ `0dad7b822255a0ae261ec45ae937371e859ffd1a` ...)
1 · Systems sources (mindspore-ai/akg @ `5aa15f3fe4d856eb60cd3b27447e04f5f1fb4dff` ...)
1 · Systems sources (sgl-project/sglang @ `16176278d0180f1a1a578e53e6eef432577876c2` ...)
1 · Systems sources (vllm-project/vllm @ `537a79f0fc5ba564d2bef43d54915b2f80f9f77d` (root ...) | candidate · not_run · 7 readset(s) | no-eval | | [optimization-search-policy](../src/inferrsi/data/skill_library/methods/optimization-search-policy/SKILL.md) | common | Policy for how an optimization agent should select, order, and stop optimization experiments. Covers candidate ... | Selecting, ordering and stopping optimization experiments: candidate generation ... | failure_modes, input_contract, produced_artifact, refusal_point
4 advice | 1 Accelerator · Optimize | 2 · Methodology and corpus references (JiusiServe/vllm-evolve @ 43cc82c)
2 · Methodology and corpus references (RRSI — arXiv 2609.24972v3)
2 · Methodology and corpus references (arXiv 2608.01867 (CRISP))
2 · Methodology and corpus references (microsoft/SkillOpt @ 343db22) | candidate · not_run · 9 readset(s) | no-eval | diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/README.md b/src/inferrsi/data/repo_packs/inferrsi/knowledge/README.md index 86ee95d7e..f2093bc6b 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/README.md +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/README.md @@ -15,7 +15,7 @@ These are knowledge abstracts (`--kind knowledge`), with namespaced package IDs distinct from explicitly selected `checkout_ref` skills. Ascend/CUDA procedures use their corresponding backend; shared methods use `common`. -Source snapshot: [7aa3344bf84c42467f43b3a927e1c7d13642bd55](https://github.com/InferMatrixLab/InferRSI/tree/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library). +Source snapshot: [bad68ff5c1426b050df78ffbd106ac2e370ce186](https://github.com/InferMatrixLab/InferRSI/tree/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library). ```bash inferrsi knowledge search 'profiling' --repo-type inferrsi --limit 50 @@ -29,7 +29,7 @@ contain the current input paths and bytes (Git checkout filters are respected). ```bash python tools/build_skill_index.py --source-revision -python tools/build_skill_index.py --source-revision 7aa3344bf84c42467f43b3a927e1c7d13642bd55 --check +python tools/build_skill_index.py --source-revision bad68ff5c1426b050df78ffbd106ac2e370ce186 --check ``` `--check` detects missing, stale and unexpected generated records without writing. diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-aclgraph-capture-replay.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-aclgraph-capture-replay.json index d628bdadd..fc4c8119c 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-aclgraph-capture-replay.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-aclgraph-capture-replay.json @@ -21,8 +21,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/ascend/ascend-aclgraph-capture-replay/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/ascend/ascend-aclgraph-capture-replay/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/ascend/ascend-aclgraph-capture-replay/SKILL.md", "src/inferrsi/data/skill_library/hardware/ascend/ascend-aclgraph-capture-replay/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-device-health.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-device-health.json index f4b372d19..f00e262f6 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-device-health.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-device-health.json @@ -23,8 +23,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/ascend/ascend-device-health/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/ascend/ascend-device-health/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/ascend/ascend-device-health/SKILL.md", "src/inferrsi/data/skill_library/hardware/ascend/ascend-device-health/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-distributed-parallel.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-distributed-parallel.json index 4b940539c..e0adde5cc 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-distributed-parallel.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-distributed-parallel.json @@ -27,8 +27,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/ascend/ascend-distributed-parallel/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/ascend/ascend-distributed-parallel/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/ascend/ascend-distributed-parallel/SKILL.md", "src/inferrsi/data/skill_library/hardware/ascend/ascend-distributed-parallel/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-graph-mode.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-graph-mode.json index 5c1c8590b..a1f70fbfd 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-graph-mode.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-graph-mode.json @@ -26,8 +26,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/ascend/ascend-graph-mode/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/ascend/ascend-graph-mode/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/ascend/ascend-graph-mode/SKILL.md", "src/inferrsi/data/skill_library/hardware/ascend/ascend-graph-mode/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-model-adaptation.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-model-adaptation.json index 971e9813a..9f306d209 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-model-adaptation.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-model-adaptation.json @@ -22,8 +22,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/ascend/ascend-model-adaptation/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/ascend/ascend-model-adaptation/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/ascend/ascend-model-adaptation/SKILL.md", "src/inferrsi/data/skill_library/hardware/ascend/ascend-model-adaptation/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-operator-profiling.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-operator-profiling.json index e8d595cbc..d4e7deaaa 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-operator-profiling.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-operator-profiling.json @@ -23,8 +23,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/ascend/ascend-operator-profiling/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/ascend/ascend-operator-profiling/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/ascend/ascend-operator-profiling/SKILL.md", "src/inferrsi/data/skill_library/hardware/ascend/ascend-operator-profiling/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-operator-tuning.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-operator-tuning.json index 64816a247..4c0841f9a 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-operator-tuning.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-operator-tuning.json @@ -24,8 +24,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/ascend/ascend-operator-tuning/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/ascend/ascend-operator-tuning/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/ascend/ascend-operator-tuning/SKILL.md", "src/inferrsi/data/skill_library/hardware/ascend/ascend-operator-tuning/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-optimization-casebook.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-optimization-casebook.json index 7521c9f69..a116bb5c3 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-optimization-casebook.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-optimization-casebook.json @@ -22,8 +22,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/ascend/ascend-optimization-casebook/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/ascend/ascend-optimization-casebook/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/ascend/ascend-optimization-casebook/SKILL.md", "src/inferrsi/data/skill_library/hardware/ascend/ascend-optimization-casebook/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-performance-analysis.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-performance-analysis.json index 2a4a0ace2..a8333140d 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-performance-analysis.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-performance-analysis.json @@ -22,8 +22,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/ascend/ascend-performance-analysis/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/ascend/ascend-performance-analysis/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/ascend/ascend-performance-analysis/SKILL.md", "src/inferrsi/data/skill_library/hardware/ascend/ascend-performance-analysis/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-video-diffusion-optimization.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-video-diffusion-optimization.json index a2d677d4d..8c2df73a1 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-video-diffusion-optimization.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascend-video-diffusion-optimization.json @@ -26,8 +26,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/ascend/ascend-video-diffusion-optimization/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/ascend/ascend-video-diffusion-optimization/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/ascend/ascend-video-diffusion-optimization/SKILL.md", "src/inferrsi/data/skill_library/hardware/ascend/ascend-video-diffusion-optimization/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascendc-tiling-design.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascendc-tiling-design.json index ea618b662..6bb94db0e 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascendc-tiling-design.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/ascendc-tiling-design.json @@ -24,8 +24,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/ascend/ascendc-tiling-design/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/ascend/ascendc-tiling-design/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/ascend/ascendc-tiling-design/SKILL.md", "src/inferrsi/data/skill_library/hardware/ascend/ascendc-tiling-design/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/cann-attribution-overlap.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/cann-attribution-overlap.json index ca740b5b7..ec809cba5 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/cann-attribution-overlap.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/cann-attribution-overlap.json @@ -24,8 +24,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/ascend/cann-attribution-overlap/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/ascend/cann-attribution-overlap/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/ascend/cann-attribution-overlap/SKILL.md", "src/inferrsi/data/skill_library/hardware/ascend/cann-attribution-overlap/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/torch-npu.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/torch-npu.json index 7c1cde5ed..9fbe96a20 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/torch-npu.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/torch-npu.json @@ -20,8 +20,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/frameworks/torch-npu/torch-npu/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/frameworks/torch-npu/torch-npu/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/frameworks/torch-npu/torch-npu/SKILL.md", "src/inferrsi/data/skill_library/frameworks/torch-npu/torch-npu/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/torchnpu-device-memory-lifecycle.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/torchnpu-device-memory-lifecycle.json index d6676b490..f49507d6d 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/torchnpu-device-memory-lifecycle.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/torchnpu-device-memory-lifecycle.json @@ -22,8 +22,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/ascend/torchnpu-device-memory-lifecycle/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/ascend/torchnpu-device-memory-lifecycle/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/ascend/torchnpu-device-memory-lifecycle/SKILL.md", "src/inferrsi/data/skill_library/hardware/ascend/torchnpu-device-memory-lifecycle/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/torchnpu-pinned-host-memory-lifecycle.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/torchnpu-pinned-host-memory-lifecycle.json index d6ea1871c..0150cca5b 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/torchnpu-pinned-host-memory-lifecycle.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/ascend/skills/torchnpu-pinned-host-memory-lifecycle.json @@ -22,8 +22,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/ascend/torchnpu-pinned-host-memory-lifecycle/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/ascend/torchnpu-pinned-host-memory-lifecycle/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/ascend/torchnpu-pinned-host-memory-lifecycle/SKILL.md", "src/inferrsi/data/skill_library/hardware/ascend/torchnpu-pinned-host-memory-lifecycle/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/assess-serving-preemption.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/assess-serving-preemption.json index 12aeca1fd..c15060f34 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/assess-serving-preemption.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/assess-serving-preemption.json @@ -16,8 +16,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/methods/assess-serving-preemption/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/methods/assess-serving-preemption/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/methods/assess-serving-preemption/SKILL.md", "src/inferrsi/data/skill_library/methods/assess-serving-preemption/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/assess-split-backward-schedule.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/assess-split-backward-schedule.json index 7c6e81473..089cdbace 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/assess-split-backward-schedule.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/assess-split-backward-schedule.json @@ -16,8 +16,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/methods/assess-split-backward-schedule/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/methods/assess-split-backward-schedule/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/methods/assess-split-backward-schedule/SKILL.md", "src/inferrsi/data/skill_library/methods/assess-split-backward-schedule/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-fixed-workload-claim.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-fixed-workload-claim.json index a3b85ccdc..c3457e9dd 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-fixed-workload-claim.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-fixed-workload-claim.json @@ -20,8 +20,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/methods/audit-fixed-workload-claim/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/methods/audit-fixed-workload-claim/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/methods/audit-fixed-workload-claim/SKILL.md", "src/inferrsi/data/skill_library/methods/audit-fixed-workload-claim/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-h3-metal-fasth3-semantics.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-h3-metal-fasth3-semantics.json index 50b90d31b..c7ddfe758 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-h3-metal-fasth3-semantics.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-h3-metal-fasth3-semantics.json @@ -29,8 +29,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/h3-metal/audit-h3-metal-fasth3-semantics/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/h3-metal/audit-h3-metal-fasth3-semantics/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/h3-metal/audit-h3-metal-fasth3-semantics/LICENSE", "src/inferrsi/data/skill_library/repositories/h3-metal/audit-h3-metal-fasth3-semantics/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-h3-runtime-residency.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-h3-runtime-residency.json index c01e0e11f..797bacfce 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-h3-runtime-residency.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-h3-runtime-residency.json @@ -30,8 +30,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/h3-c/audit-h3-runtime-residency/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/h3-c/audit-h3-runtime-residency/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/h3-c/audit-h3-runtime-residency/LICENSE", "src/inferrsi/data/skill_library/repositories/h3-c/audit-h3-runtime-residency/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-llama-runtime-state.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-llama-runtime-state.json index 3adab00f6..315ecca2a 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-llama-runtime-state.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-llama-runtime-state.json @@ -23,8 +23,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/llama-cpp/audit-llama-runtime-state/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/llama-cpp/audit-llama-runtime-state/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/llama-cpp/audit-llama-runtime-state/SKILL.md", "src/inferrsi/data/skill_library/repositories/llama-cpp/audit-llama-runtime-state/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-recurrent-kv-finalization.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-recurrent-kv-finalization.json index d5ab3d293..225088f47 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-recurrent-kv-finalization.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-recurrent-kv-finalization.json @@ -26,8 +26,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/vllm-rlt/audit-recurrent-kv-finalization/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/vllm-rlt/audit-recurrent-kv-finalization/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/vllm-rlt/audit-recurrent-kv-finalization/LICENSE", "src/inferrsi/data/skill_library/repositories/vllm-rlt/audit-recurrent-kv-finalization/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-sd-cpp-residency.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-sd-cpp-residency.json index 8b883145a..cc998ac4e 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-sd-cpp-residency.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-sd-cpp-residency.json @@ -25,8 +25,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/stable-diffusion-cpp/audit-sd-cpp-residency/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/stable-diffusion-cpp/audit-sd-cpp-residency/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/stable-diffusion-cpp/audit-sd-cpp-residency/LICENSE", "src/inferrsi/data/skill_library/repositories/stable-diffusion-cpp/audit-sd-cpp-residency/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-sglang-diffusion-lifetimes.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-sglang-diffusion-lifetimes.json index 2dab44411..82d870d72 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-sglang-diffusion-lifetimes.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-sglang-diffusion-lifetimes.json @@ -23,8 +23,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/sglang/audit-sglang-diffusion-lifetimes/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/sglang/audit-sglang-diffusion-lifetimes/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/sglang/audit-sglang-diffusion-lifetimes/LICENSE", "src/inferrsi/data/skill_library/repositories/sglang/audit-sglang-diffusion-lifetimes/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-vllm-kv-lifetimes.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-vllm-kv-lifetimes.json index a4074b797..6e713b175 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-vllm-kv-lifetimes.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/audit-vllm-kv-lifetimes.json @@ -21,8 +21,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/vllm/audit-vllm-kv-lifetimes/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/vllm/audit-vllm-kv-lifetimes/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/vllm/audit-vllm-kv-lifetimes/SKILL.md", "src/inferrsi/data/skill_library/repositories/vllm/audit-vllm-kv-lifetimes/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/bound-windowed-attention-replay.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/bound-windowed-attention-replay.json index 3c226dfe4..f2f1435e2 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/bound-windowed-attention-replay.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/bound-windowed-attention-replay.json @@ -23,8 +23,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/deepinfer/bound-windowed-attention-replay/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/deepinfer/bound-windowed-attention-replay/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/deepinfer/bound-windowed-attention-replay/SKILL.md", "src/inferrsi/data/skill_library/repositories/deepinfer/bound-windowed-attention-replay/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/budget-speculative-verification.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/budget-speculative-verification.json index bb08f2ed5..f60c205ab 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/budget-speculative-verification.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/budget-speculative-verification.json @@ -23,8 +23,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/deepinfer/budget-speculative-verification/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/deepinfer/budget-speculative-verification/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/deepinfer/budget-speculative-verification/SKILL.md", "src/inferrsi/data/skill_library/repositories/deepinfer/budget-speculative-verification/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/calibrate-serving-simulation.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/calibrate-serving-simulation.json index 3e86452fe..3c141e98a 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/calibrate-serving-simulation.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/calibrate-serving-simulation.json @@ -16,8 +16,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/methods/calibrate-serving-simulation/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/methods/calibrate-serving-simulation/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/methods/calibrate-serving-simulation/SKILL.md", "src/inferrsi/data/skill_library/methods/calibrate-serving-simulation/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/cgroup-resource-controls.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/cgroup-resource-controls.json index a88384b33..aaeb8345a 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/cgroup-resource-controls.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/cgroup-resource-controls.json @@ -25,8 +25,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/host/cgroup-resource-controls/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/host/cgroup-resource-controls/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/host/cgroup-resource-controls/SKILL.md", "src/inferrsi/data/skill_library/hardware/host/cgroup-resource-controls/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/checkpoint-recovery-consistency.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/checkpoint-recovery-consistency.json index 3572a1f8c..230499aaa 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/checkpoint-recovery-consistency.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/checkpoint-recovery-consistency.json @@ -22,8 +22,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/methods/checkpoint-recovery-consistency/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/methods/checkpoint-recovery-consistency/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/methods/checkpoint-recovery-consistency/SKILL.md", "src/inferrsi/data/skill_library/methods/checkpoint-recovery-consistency/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/choose-parallel-training-plan.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/choose-parallel-training-plan.json index d5ce0bdbf..2522b8004 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/choose-parallel-training-plan.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/choose-parallel-training-plan.json @@ -16,8 +16,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/methods/choose-parallel-training-plan/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/methods/choose-parallel-training-plan/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/methods/choose-parallel-training-plan/SKILL.md", "src/inferrsi/data/skill_library/methods/choose-parallel-training-plan/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/cite-check-downstream-pr.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/cite-check-downstream-pr.json index ded00b5e0..6eaf36df8 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/cite-check-downstream-pr.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/cite-check-downstream-pr.json @@ -22,8 +22,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/workflows/cite-check-downstream-pr/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/workflows/cite-check-downstream-pr/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/skill-index.json", "src/inferrsi/data/skill_library/workflows/cite-check-downstream-pr/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/claim-contract.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/claim-contract.json index a7f2bfa5d..5e9af4f73 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/claim-contract.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/claim-contract.json @@ -22,8 +22,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/workflows/claim-contract/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/workflows/claim-contract/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/skill-index.json", "src/inferrsi/data/skill_library/workflows/claim-contract/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/comfy-ui.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/comfy-ui.json index acdc57b97..aa7a4e1f3 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/comfy-ui.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/comfy-ui.json @@ -22,8 +22,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/comfy-ui/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/comfy-ui/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/comfy-ui/LICENSE", "src/inferrsi/data/skill_library/repositories/comfy-ui/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/competitor-scan.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/competitor-scan.json index 0f5b0fb34..68190d27d 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/competitor-scan.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/competitor-scan.json @@ -21,8 +21,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/workflows/competitor-scan/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/workflows/competitor-scan/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/skill-index.json", "src/inferrsi/data/skill_library/workflows/competitor-scan/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/dcp-async-save-performance.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/dcp-async-save-performance.json index 064e2ca25..61c877d26 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/dcp-async-save-performance.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/dcp-async-save-performance.json @@ -23,8 +23,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/frameworks/pytorch/dcp-async-save-performance/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/frameworks/pytorch/dcp-async-save-performance/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/frameworks/pytorch/dcp-async-save-performance/SKILL.md", "src/inferrsi/data/skill_library/frameworks/pytorch/dcp-async-save-performance/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/dense-performance-evidence.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/dense-performance-evidence.json index b20edd811..890d0e9b2 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/dense-performance-evidence.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/dense-performance-evidence.json @@ -21,8 +21,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/methods/dense-performance-evidence/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/methods/dense-performance-evidence/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/methods/dense-performance-evidence/SKILL.md", "src/inferrsi/data/skill_library/methods/dense-performance-evidence/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/diagnose-vllm-admission-pressure.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/diagnose-vllm-admission-pressure.json index 70fd8e8a5..73ebe1f53 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/diagnose-vllm-admission-pressure.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/diagnose-vllm-admission-pressure.json @@ -21,8 +21,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/vllm/diagnose-vllm-admission-pressure/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/vllm/diagnose-vllm-admission-pressure/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/vllm/diagnose-vllm-admission-pressure/SKILL.md", "src/inferrsi/data/skill_library/repositories/vllm/diagnose-vllm-admission-pressure/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/diffusers-invocation-state.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/diffusers-invocation-state.json index 3a883cf9a..5ac6e101b 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/diffusers-invocation-state.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/diffusers-invocation-state.json @@ -22,8 +22,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/frameworks/diffusers/diffusers-invocation-state/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/frameworks/diffusers/diffusers-invocation-state/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/frameworks/diffusers/diffusers-invocation-state/SKILL.md", "src/inferrsi/data/skill_library/frameworks/diffusers/diffusers-invocation-state/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/draft-acceptance-issue.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/draft-acceptance-issue.json index e90c2488d..5099c93e8 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/draft-acceptance-issue.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/draft-acceptance-issue.json @@ -22,8 +22,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/workflows/draft-acceptance-issue/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/workflows/draft-acceptance-issue/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/skill-index.json", "src/inferrsi/data/skill_library/workflows/draft-acceptance-issue/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/dwarfstar.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/dwarfstar.json index c456f30e9..39a016008 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/dwarfstar.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/dwarfstar.json @@ -25,8 +25,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/dwarfstar/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/dwarfstar/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/dwarfstar/LICENSE", "src/inferrsi/data/skill_library/repositories/dwarfstar/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/evaluate-elastic-serving-placement.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/evaluate-elastic-serving-placement.json index 3482d1338..c99be486d 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/evaluate-elastic-serving-placement.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/evaluate-elastic-serving-placement.json @@ -16,8 +16,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/methods/evaluate-elastic-serving-placement/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/methods/evaluate-elastic-serving-placement/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/methods/evaluate-elastic-serving-placement/SKILL.md", "src/inferrsi/data/skill_library/methods/evaluate-elastic-serving-placement/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/generation-cache-lifecycle.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/generation-cache-lifecycle.json index 7d68adf8b..e0b4a6ce9 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/generation-cache-lifecycle.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/generation-cache-lifecycle.json @@ -22,8 +22,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/mlx-lm/generation-cache-lifecycle/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/mlx-lm/generation-cache-lifecycle/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/mlx-lm/generation-cache-lifecycle/SKILL.md", "src/inferrsi/data/skill_library/repositories/mlx-lm/generation-cache-lifecycle/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/host-memory-management.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/host-memory-management.json index bb39eac1d..21b38beb6 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/host-memory-management.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/host-memory-management.json @@ -30,8 +30,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/host/host-memory-management/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/host/host-memory-management/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/host/host-memory-management/SKILL.md", "src/inferrsi/data/skill_library/hardware/host/host-memory-management/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/host-overhead-optimization.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/host-overhead-optimization.json index 9f20ba9d7..3b8e30cc9 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/host-overhead-optimization.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/host-overhead-optimization.json @@ -22,8 +22,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/host/host-overhead-optimization/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/host/host-overhead-optimization/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/host/host-overhead-optimization/SKILL.md", "src/inferrsi/data/skill_library/hardware/host/host-overhead-optimization/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/husky.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/husky.json index 3dee4b666..cb53d7046 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/husky.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/husky.json @@ -21,8 +21,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/husky/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/husky/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/husky/SKILL.md", "src/inferrsi/data/skill_library/repositories/husky/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/inference-benchmark-design.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/inference-benchmark-design.json index 3b2327374..9a3efb241 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/inference-benchmark-design.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/inference-benchmark-design.json @@ -27,7 +27,10 @@ "stream-boundaries", "treatment-delivery", "trial-validity", - "protocol-audit" + "protocol-audit", + "accepted-prefix", + "state-restoration", + "completion-accounting" ], "origin": [ "external" @@ -35,8 +38,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/methods/inference-benchmark-design/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/methods/inference-benchmark-design/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/methods/inference-benchmark-design/SKILL.md", "src/inferrsi/data/skill_library/methods/inference-benchmark-design/metadata.json", @@ -46,10 +49,12 @@ "src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/cache-aware-asr-stream-boundaries.md", "src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/justgoal-evaluation-contracts.md", "src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/justgoal-source.json", + "src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/ninfer-source.json", + "src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/ninfer-state-and-completion-contracts.md", "src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/task-benchmark-selection.md", "src/inferrsi/data/skill_library/skill-index.json" ], - "content_sha256": "7fd7f557f5c99ca4ec4c46f2d97e67ecf280f5f0fdf071871a14cc08a7cb1c84" + "content_sha256": "63926fcc15953c1a00c12c0c92fa6e4eaf9d89e52b34727eaecd720d693dd293" } ], "index_type": "packaged-skill", @@ -58,14 +63,16 @@ }, "checkout_ref": "checkout:inference-benchmark-design", "source_files": { - "src/inferrsi/data/skill_library/methods/inference-benchmark-design/SKILL.md": "7fd7f557f5c99ca4ec4c46f2d97e67ecf280f5f0fdf071871a14cc08a7cb1c84", - "src/inferrsi/data/skill_library/methods/inference-benchmark-design/metadata.json": "a646de55162cb8160f3f0947cab4b6ecac81927c3225019044b3f909087ba048", + "src/inferrsi/data/skill_library/methods/inference-benchmark-design/SKILL.md": "63926fcc15953c1a00c12c0c92fa6e4eaf9d89e52b34727eaecd720d693dd293", + "src/inferrsi/data/skill_library/methods/inference-benchmark-design/metadata.json": "8bd77f9443f700663f05adfb3ca5a707babf7bdd16784a946bd4143294d89730", "src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/autotunebench-source.json": "bc343ece2bf40c4e8aa4bd8d84f32465d00633255557feda5fe7799ce24e8688", "src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/autotunebench-trial-validity.md": "cb3a7a5d94d056f868af07f92a16a6fef31eafd3199ed07ae79e93798e320b1d", "src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/benchmark-commands.md": "0898424e0bff3a6246e1eeeda6abd6c5bd974c047fe32f3bbd5ab9d31ce17cd3", "src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/cache-aware-asr-stream-boundaries.md": "4e0ea9c1bf6b096a49a91ad5266e90b8a456f4f5aca15eba8ac270a53c2309b5", "src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/justgoal-evaluation-contracts.md": "d25df795fd301e8a1bb436e5f8122d0bac2441c43104e59a3b310fd5a64f2f3e", "src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/justgoal-source.json": "60f2784384ea1a3b08a2ecafb60207d8e3707247dc7bc177b95b463d81f1cc16", + "src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/ninfer-source.json": "ccda0fc6781e3c4dceb41521ab952445840b4afcc9f5a7d968857f76e8db6272", + "src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/ninfer-state-and-completion-contracts.md": "2b3de210d1790eec92319addadf0e38fcd1529fa2f23b5cddf96a789b783f4ac", "src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/task-benchmark-selection.md": "0d4bb080a754e26846e81680023db0dff5e6a4c96be4327c411c3b72ac4ef8e2", "src/inferrsi/data/skill_library/skill-index.json": "9ea8b2ef190777e683a3905f0614612a37782be4d655d230c000188b0c7073db" }, @@ -90,7 +97,10 @@ "stream-boundaries", "treatment-delivery", "trial-validity", - "protocol-audit" + "protocol-audit", + "accepted-prefix", + "state-restoration", + "completion-accounting" ], "status": "candidate", "profile": "local-skill-v1", @@ -310,14 +320,51 @@ "Tables 1-7" ], "readset": "references/autotunebench-trial-validity.md" + }, + { + "type": "git", + "url": "https://github.com/Neroued/ninfer", + "commit": "1882c6adbc98043fcf89f3589aa5addb3742777a", + "paths": [ + "LICENSE", + "README.md", + "docs/maintainer/engine-architecture.md", + "docs/maintainer/speculative-decoding.md", + "docs/maintainer/context-management.md", + "docs/performance/methodology.md", + "docs/performance/qwen3.8-27b.md", + "include/ninfer/ops/gdn_replay.h", + "src/core/gdn_replay_records.h", + "src/core/gdn_replay_records.cpp", + "src/core/linear_attention_state.h", + "src/core/linear_attention_state.cpp", + "src/models/qwen3_5/state/state_image.h", + "src/models/qwen3_5/state/state_image.cpp", + "src/models/qwen3_5/execution/gdn.cpp", + "src/models/qwen3_5/execution/text.cpp", + "src/models/qwen3_5/program/decode.cpp", + "src/models/qwen3_5/program/commit.cpp", + "src/models/qwen3_5/program/mtp.cpp", + "src/models/qwen3_5/session/commit.cpp", + "src/models/qwen3_5/session/binding.cpp", + "src/models/qwen3_5/session/checkpoints.cpp", + "src/runtime/engine/engine_execution.inl", + "src/runtime/executor/contract.h", + "src/runtime/executor/local_executor.h", + "src/ops/linear_attention/gated_delta_net/replay/fold.cpp", + "tests/ops/linear_attention/gated_delta_net/test_fold.cpp" + ], + "readset": "references/ninfer-source.json", + "license": "Apache-2.0", + "scope": "Selected Qwen3.5 GDN/MTP state-restoration, accepted-prefix and completion evidence contracts; historical results retain their separate reported revision. Source-only, no native execution." } ], "empirical_validation": "not_run", "applicability": { - "scope": "Selecting task quality gates, reference parity checks and performance workloads for text, agentic coding, work agents, cache-aware ASR, TTS, full-duplex audio/video, image and video inference, and designing goodput-based multi-regime comparisons that fit InferRSI's evidence trail; interpreting treatment delivery, arm validity and stated-versus-observed protocol enforcement in supplied tuning reports", + "scope": "Selecting task quality gates, reference parity checks and performance workloads for text, agentic coding, work agents, cache-aware ASR, TTS, full-duplex audio/video, image and video inference, and designing goodput-based multi-regime comparisons that fit InferRSI's evidence trail; interpreting treatment delivery, arm validity and stated-versus-observed protocol enforcement in supplied tuning reports; designing and interpreting the selected NInfer hybrid speculative state-restoration and completion validation contract", "prerequisites": "The primary objective, timing boundary, quality gates, sample/failure accounting, required regimes, repetitions, seeds and decision rule are frozen before optimization; baseline and candidate share hardware, software stack, input distribution, benchmark configuration, warmup and measurement protocol, and on a shared machine also one time window and dedicated devices; cache-aware ASR also requires the selected model/config and first/steady geometry, decoded clock and reset/EOU/EOF boundaries to be established", - "input_contract": "The task family and deployment use case; candidate quality evaluators and reference parity; the declared SLO and the caller's policy on which regimes must improve or must not regress; for an audit, a supplied fixed-workload report with request-to-trial grouping, exclusion and failure counts and the declared aggregation rule; for a treatment audit, the frozen delivery/arm-validity rules, run/candidate identities, feedback and consumer-input evidence, and per-arm raw records", - "produced_artifact": "A frozen benchmark design with evaluators, reference parity, primary objective, timing boundary, quality gates, required regimes, repetition and seed rules and an explicit decision config, plus the evidence-tree layout a PASS claim must produce; for a supplied report, the computed threshold result and missing evidence. Candidate knowledge, not qualification of any model or backend; for a treatment audit, a contract-bound eligibility and interpretation report separating treatment exposure, observed checks, comparator identity, anomaly disposition and source conflicts", + "input_contract": "The task family and deployment use case; candidate quality evaluators and reference parity; the declared SLO and the caller's policy on which regimes must improve or must not regress; for an audit, a supplied fixed-workload report with request-to-trial grouping, exclusion and failure counts and the declared aggregation rule; for a treatment audit, the frozen delivery/arm-validity rules, run/candidate identities, feedback and consumer-input evidence, and per-arm raw records; for the NInfer slice, a complete checkpoint/frontier identity, physical verify-block/state oracle, final adopted extents and correlated execution/termination records", + "produced_artifact": "A frozen benchmark design with evaluators, reference parity, primary objective, timing boundary, quality gates, required regimes, repetition and seed rules and an explicit decision config, plus the evidence-tree layout a PASS claim must produce; for a supplied report, the computed threshold result and missing evidence. Candidate knowledge, not qualification of any model or backend; for a treatment audit, a contract-bound eligibility and interpretation report separating treatment exposure, observed checks, comparator identity, anomaly disposition and source conflicts; for the NInfer slice, a source-bound validation plan or report interpretation that separates state-oracle evidence, final adoption, retirement and workload denominators", "failure_modes": [ "Raw throughput is optimized in place of goodput, so requests served outside the SLO count toward a win", "One concurrency level or one undisclosed aggregate score stands in for all required regimes", @@ -329,16 +376,19 @@ "A cache-aware ASR EOU is treated as EOF or full reset, nominal chunk duration as the decoded mel frontier, or shape-only geometry PASS as a native model-state test", "An enabled treatment or stored feedback summary is taken as proof of delivery to the next eligible consumer", "A failed arm-validity condition is overridden by a favorable best timing or a saturation/null interpretation", - "Complete gate keys or zero recorded failures are read as proof that every declared protocol check ran" + "Complete gate keys or zero recorded failures are read as proof that every declared protocol check ran", + "Zero accepted drafts is treated as zero adopted input, or a matching token prefix is used to splice unrelated State and KV checkpoints", + "ResultReady or a commit receipt is treated as DeviceRetired, or selected full-batch intervals are compared with complete-cohort rates under one unlabeled denominator" ], - "refusal_point": "Records missing coverage and high-variance results as INCONCLUSIVE rather than a win; a run without a frozen RunSpec, or whose validator changed after the plan was frozen, is INCONCLUSIVE regardless of metric values; sets no universal numeric gate or noise threshold; withholds causal treatment benefit/null conclusions when required delivery or arm eligibility is unverified, and preserves conflicts between stated protocol, exported code and source reports" + "refusal_point": "Records missing coverage and high-variance results as INCONCLUSIVE rather than a win; a run without a frozen RunSpec, or whose validator changed after the plan was frozen, is INCONCLUSIVE regardless of metric values; sets no universal numeric gate or noise threshold; withholds causal treatment benefit/null conclusions when required delivery or arm eligibility is unverified, and preserves conflicts between stated protocol, exported code and source reports; withholds native state-correctness and cancellation/reuse conclusions when the required same-block oracle, checkpoint identity or matching receipt is missing" }, "recommendations": [ "Freeze one primary objective, timing boundary, quality gates and sample/failure accounting before optimization.", "Choose the required regimes from the deployment use case, report each separately, and record which must improve or must not regress in a versioned decision config.", "Use the replay trace that matches the actual deployment pattern.", "State the concurrency, trace, model, SKU and microbenchmark scope boundaries in any report that cites benchmark results.", - "Trace treatment to its eligible consumer and apply the frozen arm-validity precedence before attributing benefit, saturation or a null." + "Trace treatment to its eligible consumer and apply the frozen arm-validity precedence before attributing benefit, saturation or a null.", + "For hybrid speculative comparisons, freeze the coupled restore point, same-physical-block oracle, zero-draft/zero-adoption controls and complete-cohort versus steady-phase denominators before interpreting a result." ], "limitations": [ "Source-derived candidate; no local efficacy validation.", @@ -347,7 +397,8 @@ "Benchmark feasibility and package-consumption checks do not promote this candidate or replace task quality, repeated A/B and service qualification.", "The pinned NeMo-Speech.cpp ASR reference covers selected source/test definitions and author-reported benchmark boundaries; native model/device, decoder internals, quality, long-audio/load and service behavior remain not_run.", "JustGoal contributes source-only controlled-evaluation and kernel-acceptance guidance; its repaired/retested timings, released-pass counts and protocol differences are author-reported, not locally reproduced or a complete campaign runner.", - "The AutoTuneBench reference is source-only: original operational checkpoint/prompt receipts remain unavailable, source contradictions remain explicit, and no tuning, device/performance or agent-efficacy validation was run." + "The AutoTuneBench reference is source-only: original operational checkpoint/prompt receipts remain unavailable, source contradictions remain explicit, and no tuning, device/performance or agent-efficacy validation was run.", + "The NInfer reference inspects selected Qwen3.5 GDN/MTP source contracts and unexecuted test definitions; its historical author measurements use a different engine revision. Native state/model/device/quality/performance and efficacy remain unvalidated." ], "provenance_readsets": [ "references/task-benchmark-selection.md", @@ -356,20 +407,24 @@ "references/justgoal-evaluation-contracts.md", "references/justgoal-source.json", "references/autotunebench-trial-validity.md", - "references/autotunebench-source.json" + "references/autotunebench-source.json", + "references/ninfer-state-and-completion-contracts.md", + "references/ninfer-source.json" ], "references": [ "../inference-optimization-casebook/cases/admission-drain-occupancy.md", - "../inference-optimization-casebook/cases/admission-drain-occupancy.json" + "../inference-optimization-casebook/cases/admission-drain-occupancy.json", + "../../repositories/deepinfer/budget-speculative-verification/SKILL.md", + "../../repositories/vllm-rlt/audit-recurrent-kv-finalization/SKILL.md" ], "consumption_mode": "packaged", "searchspace_not_applicable": "SKILL.md §§ Select the task and its evaluation contract, Statistical requirements, Shared-machine measurement discipline, and Search integrity and final-state verification design comparable evaluation; hardware identity and allocation isolation are comparison controls, not a hardware-resource mechanism." }, "applicability": { - "scope": "Selecting task quality gates, reference parity checks and performance workloads for text, agentic coding, work agents, cache-aware ASR, TTS, full-duplex audio/video, image and video inference, and designing goodput-based multi-regime comparisons that fit InferRSI's evidence trail; interpreting treatment delivery, arm validity and stated-versus-observed protocol enforcement in supplied tuning reports", + "scope": "Selecting task quality gates, reference parity checks and performance workloads for text, agentic coding, work agents, cache-aware ASR, TTS, full-duplex audio/video, image and video inference, and designing goodput-based multi-regime comparisons that fit InferRSI's evidence trail; interpreting treatment delivery, arm validity and stated-versus-observed protocol enforcement in supplied tuning reports; designing and interpreting the selected NInfer hybrid speculative state-restoration and completion validation contract", "prerequisites": "The primary objective, timing boundary, quality gates, sample/failure accounting, required regimes, repetitions, seeds and decision rule are frozen before optimization; baseline and candidate share hardware, software stack, input distribution, benchmark configuration, warmup and measurement protocol, and on a shared machine also one time window and dedicated devices; cache-aware ASR also requires the selected model/config and first/steady geometry, decoded clock and reset/EOU/EOF boundaries to be established", - "input_contract": "The task family and deployment use case; candidate quality evaluators and reference parity; the declared SLO and the caller's policy on which regimes must improve or must not regress; for an audit, a supplied fixed-workload report with request-to-trial grouping, exclusion and failure counts and the declared aggregation rule; for a treatment audit, the frozen delivery/arm-validity rules, run/candidate identities, feedback and consumer-input evidence, and per-arm raw records", - "produced_artifact": "A frozen benchmark design with evaluators, reference parity, primary objective, timing boundary, quality gates, required regimes, repetition and seed rules and an explicit decision config, plus the evidence-tree layout a PASS claim must produce; for a supplied report, the computed threshold result and missing evidence. Candidate knowledge, not qualification of any model or backend; for a treatment audit, a contract-bound eligibility and interpretation report separating treatment exposure, observed checks, comparator identity, anomaly disposition and source conflicts", + "input_contract": "The task family and deployment use case; candidate quality evaluators and reference parity; the declared SLO and the caller's policy on which regimes must improve or must not regress; for an audit, a supplied fixed-workload report with request-to-trial grouping, exclusion and failure counts and the declared aggregation rule; for a treatment audit, the frozen delivery/arm-validity rules, run/candidate identities, feedback and consumer-input evidence, and per-arm raw records; for the NInfer slice, a complete checkpoint/frontier identity, physical verify-block/state oracle, final adopted extents and correlated execution/termination records", + "produced_artifact": "A frozen benchmark design with evaluators, reference parity, primary objective, timing boundary, quality gates, required regimes, repetition and seed rules and an explicit decision config, plus the evidence-tree layout a PASS claim must produce; for a supplied report, the computed threshold result and missing evidence. Candidate knowledge, not qualification of any model or backend; for a treatment audit, a contract-bound eligibility and interpretation report separating treatment exposure, observed checks, comparator identity, anomaly disposition and source conflicts; for the NInfer slice, a source-bound validation plan or report interpretation that separates state-oracle evidence, final adoption, retirement and workload denominators", "failure_modes": [ "Raw throughput is optimized in place of goodput, so requests served outside the SLO count toward a win", "One concurrency level or one undisclosed aggregate score stands in for all required regimes", @@ -381,16 +436,19 @@ "A cache-aware ASR EOU is treated as EOF or full reset, nominal chunk duration as the decoded mel frontier, or shape-only geometry PASS as a native model-state test", "An enabled treatment or stored feedback summary is taken as proof of delivery to the next eligible consumer", "A failed arm-validity condition is overridden by a favorable best timing or a saturation/null interpretation", - "Complete gate keys or zero recorded failures are read as proof that every declared protocol check ran" + "Complete gate keys or zero recorded failures are read as proof that every declared protocol check ran", + "Zero accepted drafts is treated as zero adopted input, or a matching token prefix is used to splice unrelated State and KV checkpoints", + "ResultReady or a commit receipt is treated as DeviceRetired, or selected full-batch intervals are compared with complete-cohort rates under one unlabeled denominator" ], - "refusal_point": "Records missing coverage and high-variance results as INCONCLUSIVE rather than a win; a run without a frozen RunSpec, or whose validator changed after the plan was frozen, is INCONCLUSIVE regardless of metric values; sets no universal numeric gate or noise threshold; withholds causal treatment benefit/null conclusions when required delivery or arm eligibility is unverified, and preserves conflicts between stated protocol, exported code and source reports" + "refusal_point": "Records missing coverage and high-variance results as INCONCLUSIVE rather than a win; a run without a frozen RunSpec, or whose validator changed after the plan was frozen, is INCONCLUSIVE regardless of metric values; sets no universal numeric gate or noise threshold; withholds causal treatment benefit/null conclusions when required delivery or arm eligibility is unverified, and preserves conflicts between stated protocol, exported code and source reports; withholds native state-correctness and cancellation/reuse conclusions when the required same-block oracle, checkpoint identity or matching receipt is missing" }, "recommendations": [ "Freeze one primary objective, timing boundary, quality gates and sample/failure accounting before optimization.", "Choose the required regimes from the deployment use case, report each separately, and record which must improve or must not regress in a versioned decision config.", "Use the replay trace that matches the actual deployment pattern.", "State the concurrency, trace, model, SKU and microbenchmark scope boundaries in any report that cites benchmark results.", - "Trace treatment to its eligible consumer and apply the frozen arm-validity precedence before attributing benefit, saturation or a null." + "Trace treatment to its eligible consumer and apply the frozen arm-validity precedence before attributing benefit, saturation or a null.", + "For hybrid speculative comparisons, freeze the coupled restore point, same-physical-block oracle, zero-draft/zero-adoption controls and complete-cohort versus steady-phase denominators before interpreting a result." ], "limitations": [ "This generated abstract points to the complete packaged skill; external evidence and upstream repositories remain source links.", @@ -403,7 +461,8 @@ "Benchmark feasibility and package-consumption checks do not promote this candidate or replace task quality, repeated A/B and service qualification.", "The pinned NeMo-Speech.cpp ASR reference covers selected source/test definitions and author-reported benchmark boundaries; native model/device, decoder internals, quality, long-audio/load and service behavior remain not_run.", "JustGoal contributes source-only controlled-evaluation and kernel-acceptance guidance; its repaired/retested timings, released-pass counts and protocol differences are author-reported, not locally reproduced or a complete campaign runner.", - "The AutoTuneBench reference is source-only: original operational checkpoint/prompt receipts remain unavailable, source contradictions remain explicit, and no tuning, device/performance or agent-efficacy validation was run." + "The AutoTuneBench reference is source-only: original operational checkpoint/prompt receipts remain unavailable, source contradictions remain explicit, and no tuning, device/performance or agent-efficacy validation was run.", + "The NInfer reference inspects selected Qwen3.5 GDN/MTP source contracts and unexecuted test definitions; its historical author measurements use a different engine revision. Native state/model/device/quality/performance and efficacy remain unvalidated." ], "verification": { "locally_verified": false, diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/inference-optimization-casebook.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/inference-optimization-casebook.json index 3d6372c43..0b3e87f47 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/inference-optimization-casebook.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/inference-optimization-casebook.json @@ -21,8 +21,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/methods/inference-optimization-casebook/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/methods/inference-optimization-casebook/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/methods/inference-optimization-casebook/SKILL.md", "src/inferrsi/data/skill_library/methods/inference-optimization-casebook/cases/admission-drain-occupancy.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/kernel-writing-guide.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/kernel-writing-guide.json index 1040189df..f8ccb1421 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/kernel-writing-guide.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/kernel-writing-guide.json @@ -39,8 +39,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/methods/kernel-writing-guide/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/methods/kernel-writing-guide/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/methods/kernel-writing-guide/SKILL.md", "src/inferrsi/data/skill_library/methods/kernel-writing-guide/cases/causal-conv1d-dispatch-state.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/kv-tier-management.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/kv-tier-management.json index 1c1f0ac80..94546efa7 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/kv-tier-management.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/kv-tier-management.json @@ -24,8 +24,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/lmcache/kv-tier-management/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/lmcache/kv-tier-management/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/lmcache/kv-tier-management/SKILL.md", "src/inferrsi/data/skill_library/repositories/lmcache/kv-tier-management/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/lazy-evaluation-stream-lifetimes.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/lazy-evaluation-stream-lifetimes.json index c3b8b61d3..6641c782f 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/lazy-evaluation-stream-lifetimes.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/lazy-evaluation-stream-lifetimes.json @@ -22,8 +22,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/frameworks/mlx/lazy-evaluation-stream-lifetimes/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/frameworks/mlx/lazy-evaluation-stream-lifetimes/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/frameworks/mlx/lazy-evaluation-stream-lifetimes/SKILL.md", "src/inferrsi/data/skill_library/frameworks/mlx/lazy-evaluation-stream-lifetimes/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/lily.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/lily.json index 8c592e80d..09a4e263c 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/lily.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/lily.json @@ -25,8 +25,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/lily/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/lily/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/lily/LICENSE", "src/inferrsi/data/skill_library/repositories/lily/NOTICE.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/lithos-metal.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/lithos-metal.json index 26722f7f4..b1c5e68be 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/lithos-metal.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/lithos-metal.json @@ -21,8 +21,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/lithos-metal/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/lithos-metal/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/lithos-metal/LICENSE", "src/inferrsi/data/skill_library/repositories/lithos-metal/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/local-dream.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/local-dream.json index cb94c8a24..66ec494fe 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/local-dream.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/local-dream.json @@ -29,8 +29,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/local-dream/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/local-dream/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/local-dream/SKILL.md", "src/inferrsi/data/skill_library/repositories/local-dream/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/mistral-rs.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/mistral-rs.json index 636fbe213..714911c30 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/mistral-rs.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/mistral-rs.json @@ -26,8 +26,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/mistral-rs/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/mistral-rs/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/mistral-rs/LICENSE", "src/inferrsi/data/skill_library/repositories/mistral-rs/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/mlc-llm.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/mlc-llm.json index b94c5f561..ec375a9ff 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/mlc-llm.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/mlc-llm.json @@ -27,8 +27,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/mlc-llm/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/mlc-llm/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/mlc-llm/LICENSE", "src/inferrsi/data/skill_library/repositories/mlc-llm/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/mlx-swift-lm.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/mlx-swift-lm.json index fef01f452..69f41859c 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/mlx-swift-lm.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/mlx-swift-lm.json @@ -27,8 +27,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/mlx-swift-lm/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/mlx-swift-lm/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/mlx-swift-lm/LICENSE", "src/inferrsi/data/skill_library/repositories/mlx-swift-lm/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/npu-skill-factory.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/npu-skill-factory.json index 1adcbdc00..66ae8afcd 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/npu-skill-factory.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/npu-skill-factory.json @@ -22,8 +22,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/workflows/npu-skill-factory/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/workflows/npu-skill-factory/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/skill-index.json", "src/inferrsi/data/skill_library/workflows/npu-skill-factory/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/optimization-search-policy.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/optimization-search-policy.json index 8965fb642..ae015529a 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/optimization-search-policy.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/optimization-search-policy.json @@ -21,8 +21,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/methods/optimization-search-policy/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/methods/optimization-search-policy/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/methods/optimization-search-policy/SKILL.md", "src/inferrsi/data/skill_library/methods/optimization-search-policy/cases/sarathi-chunk-budget.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/os-process-placement.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/os-process-placement.json index 8c0cf7b12..d85aac39c 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/os-process-placement.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/os-process-placement.json @@ -24,8 +24,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/host/os-process-placement/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/host/os-process-placement/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/host/os-process-placement/SKILL.md", "src/inferrsi/data/skill_library/hardware/host/os-process-placement/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/performance-ceiling-analysis.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/performance-ceiling-analysis.json index ae1810d82..dbb12a38c 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/performance-ceiling-analysis.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/performance-ceiling-analysis.json @@ -22,8 +22,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/methods/performance-ceiling-analysis/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/methods/performance-ceiling-analysis/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/methods/performance-ceiling-analysis/SKILL.md", "src/inferrsi/data/skill_library/methods/performance-ceiling-analysis/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/plan-activation-residency.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/plan-activation-residency.json index 5ccb28ffe..51f6b54a2 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/plan-activation-residency.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/plan-activation-residency.json @@ -16,8 +16,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/methods/plan-activation-residency/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/methods/plan-activation-residency/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/methods/plan-activation-residency/SKILL.md", "src/inferrsi/data/skill_library/methods/plan-activation-residency/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/project-maya.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/project-maya.json index abec7ae74..651860877 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/project-maya.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/project-maya.json @@ -25,8 +25,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/project-maya/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/project-maya/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/project-maya/LICENSE", "src/inferrsi/data/skill_library/repositories/project-maya/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/pytorch.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/pytorch.json index 97d160bec..7b4fc0392 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/pytorch.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/pytorch.json @@ -24,8 +24,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/frameworks/pytorch/pytorch/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/frameworks/pytorch/pytorch/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/frameworks/pytorch/pytorch/LICENSE", "src/inferrsi/data/skill_library/frameworks/pytorch/pytorch/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/repository-evolution-analysis.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/repository-evolution-analysis.json index 27968f7c8..a32a8e9ef 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/repository-evolution-analysis.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/repository-evolution-analysis.json @@ -21,8 +21,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/methods/repository-evolution-analysis/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/methods/repository-evolution-analysis/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/methods/repository-evolution-analysis/SKILL.md", "src/inferrsi/data/skill_library/methods/repository-evolution-analysis/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/rsi-evidence.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/rsi-evidence.json index 465739baa..df62a7b8b 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/rsi-evidence.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/rsi-evidence.json @@ -22,8 +22,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/workflows/rsi-evidence/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/workflows/rsi-evidence/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/skill-index.json", "src/inferrsi/data/skill_library/workflows/rsi-evidence/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/skill-evolution-playbook.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/skill-evolution-playbook.json index c470e93b8..e05302794 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/skill-evolution-playbook.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/skill-evolution-playbook.json @@ -21,8 +21,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/methods/skill-evolution-playbook/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/methods/skill-evolution-playbook/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/methods/skill-evolution-playbook/SKILL.md", "src/inferrsi/data/skill_library/methods/skill-evolution-playbook/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/splash.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/splash.json index a9476f7cf..f12ba2dbe 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/splash.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/splash.json @@ -25,8 +25,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/splash/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/splash/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/splash/LICENSE", "src/inferrsi/data/skill_library/repositories/splash/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/storage-io-measurement.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/storage-io-measurement.json index fc715e09f..898c6324e 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/storage-io-measurement.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/storage-io-measurement.json @@ -24,8 +24,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/host/storage-io-measurement/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/host/storage-io-measurement/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/host/storage-io-measurement/SKILL.md", "src/inferrsi/data/skill_library/hardware/host/storage-io-measurement/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/training-input-pipeline.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/training-input-pipeline.json index 045b18bc4..b897c72ce 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/training-input-pipeline.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/training-input-pipeline.json @@ -23,8 +23,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/methods/training-input-pipeline/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/methods/training-input-pipeline/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/methods/training-input-pipeline/SKILL.md", "src/inferrsi/data/skill_library/methods/training-input-pipeline/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/training-resource-budget.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/training-resource-budget.json index a2429cac2..add8d9946 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/training-resource-budget.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/training-resource-budget.json @@ -23,8 +23,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/methods/training-resource-budget/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/methods/training-resource-budget/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/methods/training-resource-budget/SKILL.md", "src/inferrsi/data/skill_library/methods/training-resource-budget/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/video-diffusion-optimization.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/video-diffusion-optimization.json index 2fbb6eebc..d0f70e965 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/video-diffusion-optimization.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/video-diffusion-optimization.json @@ -26,8 +26,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/methods/video-diffusion-optimization/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/methods/video-diffusion-optimization/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/methods/video-diffusion-optimization/SKILL.md", "src/inferrsi/data/skill_library/methods/video-diffusion-optimization/assets/campaign.template.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/vllm-metal.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/vllm-metal.json index 179f342ea..d675aba73 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/vllm-metal.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/vllm-metal.json @@ -26,8 +26,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/vllm-metal/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/vllm-metal/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/vllm-metal/LICENSE", "src/inferrsi/data/skill_library/repositories/vllm-metal/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/weight-loading-optimization.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/weight-loading-optimization.json index 3cc50bdee..64227fe8e 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/weight-loading-optimization.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/common/skills/weight-loading-optimization.json @@ -29,8 +29,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/methods/weight-loading-optimization/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/methods/weight-loading-optimization/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/methods/weight-loading-optimization/SKILL.md", "src/inferrsi/data/skill_library/methods/weight-loading-optimization/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/cuda/skills/audit-afd-p2p-exchanges.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/cuda/skills/audit-afd-p2p-exchanges.json index 42b227ef5..2c85886c5 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/cuda/skills/audit-afd-p2p-exchanges.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/cuda/skills/audit-afd-p2p-exchanges.json @@ -23,8 +23,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/afd-plugin/audit-afd-p2p-exchanges/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/afd-plugin/audit-afd-p2p-exchanges/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/afd-plugin/audit-afd-p2p-exchanges/SKILL.md", "src/inferrsi/data/skill_library/repositories/afd-plugin/audit-afd-p2p-exchanges/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/cuda/skills/audit-flare-session-lifecycle.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/cuda/skills/audit-flare-session-lifecycle.json index 0fc99822c..b7c945ab0 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/cuda/skills/audit-flare-session-lifecycle.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/cuda/skills/audit-flare-session-lifecycle.json @@ -29,8 +29,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/repositories/flare-local-inference/audit-flare-session-lifecycle/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/repositories/flare-local-inference/audit-flare-session-lifecycle/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/repositories/flare-local-inference/audit-flare-session-lifecycle/LICENSE", "src/inferrsi/data/skill_library/repositories/flare-local-inference/audit-flare-session-lifecycle/SKILL.md", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/cuda/skills/diagnose-attention-bottlenecks.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/cuda/skills/diagnose-attention-bottlenecks.json index 19f847d38..6c207f767 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/cuda/skills/diagnose-attention-bottlenecks.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/cuda/skills/diagnose-attention-bottlenecks.json @@ -16,8 +16,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/cuda/diagnose-attention-bottlenecks/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/cuda/diagnose-attention-bottlenecks/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/cuda/diagnose-attention-bottlenecks/SKILL.md", "src/inferrsi/data/skill_library/hardware/cuda/diagnose-attention-bottlenecks/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/cuda/skills/host-overhead-profiling.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/cuda/skills/host-overhead-profiling.json index 7fac0bc38..ea67e9be7 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/cuda/skills/host-overhead-profiling.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/cuda/skills/host-overhead-profiling.json @@ -22,8 +22,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/cuda/host-overhead-profiling/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/cuda/host-overhead-profiling/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/cuda/host-overhead-profiling/SKILL.md", "src/inferrsi/data/skill_library/hardware/cuda/host-overhead-profiling/metadata.json", diff --git a/src/inferrsi/data/repo_packs/inferrsi/knowledge/cuda/skills/reason-hopper-kernels.json b/src/inferrsi/data/repo_packs/inferrsi/knowledge/cuda/skills/reason-hopper-kernels.json index 55040913f..c5650a981 100644 --- a/src/inferrsi/data/repo_packs/inferrsi/knowledge/cuda/skills/reason-hopper-kernels.json +++ b/src/inferrsi/data/repo_packs/inferrsi/knowledge/cuda/skills/reason-hopper-kernels.json @@ -36,8 +36,8 @@ "sources": [ { "type": "git", - "url": "https://github.com/InferMatrixLab/InferRSI/blob/7aa3344bf84c42467f43b3a927e1c7d13642bd55/src/inferrsi/data/skill_library/hardware/cuda/reason-hopper-kernels/SKILL.md", - "commit": "7aa3344bf84c42467f43b3a927e1c7d13642bd55", + "url": "https://github.com/InferMatrixLab/InferRSI/blob/bad68ff5c1426b050df78ffbd106ac2e370ce186/src/inferrsi/data/skill_library/hardware/cuda/reason-hopper-kernels/SKILL.md", + "commit": "bad68ff5c1426b050df78ffbd106ac2e370ce186", "paths": [ "src/inferrsi/data/skill_library/hardware/cuda/reason-hopper-kernels/SKILL.md", "src/inferrsi/data/skill_library/hardware/cuda/reason-hopper-kernels/metadata.json", diff --git a/src/inferrsi/data/skill_library/methods/inference-benchmark-design/SKILL.md b/src/inferrsi/data/skill_library/methods/inference-benchmark-design/SKILL.md index b40a52d2c..fa2cac0f9 100644 --- a/src/inferrsi/data/skill_library/methods/inference-benchmark-design/SKILL.md +++ b/src/inferrsi/data/skill_library/methods/inference-benchmark-design/SKILL.md @@ -29,6 +29,13 @@ For treatment-enabled tuning comparisons or claimed protocol integrity, read Trace feedback to the eligible consumer input and apply the frozen arm-validity precedence before interpreting benefit, saturation or a null. +For hybrid speculative state-restoration or completion evidence, read the +[NInfer validation contract](references/ninfer-state-and-completion-contracts.md). +Freeze a coupled checkpoint and same-physical-block state oracle; distinguish +zero accepted drafts from zero adopted inputs, logical commit from device +retirement, and selected steady intervals from complete-cohort denominators. +Its source contracts and historical author reports are not native parity results. + Freeze one primary objective, timing boundary, quality gates and sample/failure accounting before optimization. Recorded agent task scores and replay completion do not establish current task success; a VBench prompt workload does not compute diff --git a/src/inferrsi/data/skill_library/methods/inference-benchmark-design/metadata.json b/src/inferrsi/data/skill_library/methods/inference-benchmark-design/metadata.json index e6ea1ff3b..20762fb58 100644 --- a/src/inferrsi/data/skill_library/methods/inference-benchmark-design/metadata.json +++ b/src/inferrsi/data/skill_library/methods/inference-benchmark-design/metadata.json @@ -17,7 +17,10 @@ "stream-boundaries", "treatment-delivery", "trial-validity", - "protocol-audit" + "protocol-audit", + "accepted-prefix", + "state-restoration", + "completion-accounting" ], "status": "candidate", "profile": "local-skill-v1", @@ -237,14 +240,51 @@ "Tables 1-7" ], "readset": "references/autotunebench-trial-validity.md" + }, + { + "type": "git", + "url": "https://github.com/Neroued/ninfer", + "commit": "1882c6adbc98043fcf89f3589aa5addb3742777a", + "paths": [ + "LICENSE", + "README.md", + "docs/maintainer/engine-architecture.md", + "docs/maintainer/speculative-decoding.md", + "docs/maintainer/context-management.md", + "docs/performance/methodology.md", + "docs/performance/qwen3.8-27b.md", + "include/ninfer/ops/gdn_replay.h", + "src/core/gdn_replay_records.h", + "src/core/gdn_replay_records.cpp", + "src/core/linear_attention_state.h", + "src/core/linear_attention_state.cpp", + "src/models/qwen3_5/state/state_image.h", + "src/models/qwen3_5/state/state_image.cpp", + "src/models/qwen3_5/execution/gdn.cpp", + "src/models/qwen3_5/execution/text.cpp", + "src/models/qwen3_5/program/decode.cpp", + "src/models/qwen3_5/program/commit.cpp", + "src/models/qwen3_5/program/mtp.cpp", + "src/models/qwen3_5/session/commit.cpp", + "src/models/qwen3_5/session/binding.cpp", + "src/models/qwen3_5/session/checkpoints.cpp", + "src/runtime/engine/engine_execution.inl", + "src/runtime/executor/contract.h", + "src/runtime/executor/local_executor.h", + "src/ops/linear_attention/gated_delta_net/replay/fold.cpp", + "tests/ops/linear_attention/gated_delta_net/test_fold.cpp" + ], + "readset": "references/ninfer-source.json", + "license": "Apache-2.0", + "scope": "Selected Qwen3.5 GDN/MTP state-restoration, accepted-prefix and completion evidence contracts; historical results retain their separate reported revision. Source-only, no native execution." } ], "empirical_validation": "not_run", "applicability": { - "scope": "Selecting task quality gates, reference parity checks and performance workloads for text, agentic coding, work agents, cache-aware ASR, TTS, full-duplex audio/video, image and video inference, and designing goodput-based multi-regime comparisons that fit InferRSI's evidence trail; interpreting treatment delivery, arm validity and stated-versus-observed protocol enforcement in supplied tuning reports", + "scope": "Selecting task quality gates, reference parity checks and performance workloads for text, agentic coding, work agents, cache-aware ASR, TTS, full-duplex audio/video, image and video inference, and designing goodput-based multi-regime comparisons that fit InferRSI's evidence trail; interpreting treatment delivery, arm validity and stated-versus-observed protocol enforcement in supplied tuning reports; designing and interpreting the selected NInfer hybrid speculative state-restoration and completion validation contract", "prerequisites": "The primary objective, timing boundary, quality gates, sample/failure accounting, required regimes, repetitions, seeds and decision rule are frozen before optimization; baseline and candidate share hardware, software stack, input distribution, benchmark configuration, warmup and measurement protocol, and on a shared machine also one time window and dedicated devices; cache-aware ASR also requires the selected model/config and first/steady geometry, decoded clock and reset/EOU/EOF boundaries to be established", - "input_contract": "The task family and deployment use case; candidate quality evaluators and reference parity; the declared SLO and the caller's policy on which regimes must improve or must not regress; for an audit, a supplied fixed-workload report with request-to-trial grouping, exclusion and failure counts and the declared aggregation rule; for a treatment audit, the frozen delivery/arm-validity rules, run/candidate identities, feedback and consumer-input evidence, and per-arm raw records", - "produced_artifact": "A frozen benchmark design with evaluators, reference parity, primary objective, timing boundary, quality gates, required regimes, repetition and seed rules and an explicit decision config, plus the evidence-tree layout a PASS claim must produce; for a supplied report, the computed threshold result and missing evidence. Candidate knowledge, not qualification of any model or backend; for a treatment audit, a contract-bound eligibility and interpretation report separating treatment exposure, observed checks, comparator identity, anomaly disposition and source conflicts", + "input_contract": "The task family and deployment use case; candidate quality evaluators and reference parity; the declared SLO and the caller's policy on which regimes must improve or must not regress; for an audit, a supplied fixed-workload report with request-to-trial grouping, exclusion and failure counts and the declared aggregation rule; for a treatment audit, the frozen delivery/arm-validity rules, run/candidate identities, feedback and consumer-input evidence, and per-arm raw records; for the NInfer slice, a complete checkpoint/frontier identity, physical verify-block/state oracle, final adopted extents and correlated execution/termination records", + "produced_artifact": "A frozen benchmark design with evaluators, reference parity, primary objective, timing boundary, quality gates, required regimes, repetition and seed rules and an explicit decision config, plus the evidence-tree layout a PASS claim must produce; for a supplied report, the computed threshold result and missing evidence. Candidate knowledge, not qualification of any model or backend; for a treatment audit, a contract-bound eligibility and interpretation report separating treatment exposure, observed checks, comparator identity, anomaly disposition and source conflicts; for the NInfer slice, a source-bound validation plan or report interpretation that separates state-oracle evidence, final adoption, retirement and workload denominators", "failure_modes": [ "Raw throughput is optimized in place of goodput, so requests served outside the SLO count toward a win", "One concurrency level or one undisclosed aggregate score stands in for all required regimes", @@ -256,16 +296,19 @@ "A cache-aware ASR EOU is treated as EOF or full reset, nominal chunk duration as the decoded mel frontier, or shape-only geometry PASS as a native model-state test", "An enabled treatment or stored feedback summary is taken as proof of delivery to the next eligible consumer", "A failed arm-validity condition is overridden by a favorable best timing or a saturation/null interpretation", - "Complete gate keys or zero recorded failures are read as proof that every declared protocol check ran" + "Complete gate keys or zero recorded failures are read as proof that every declared protocol check ran", + "Zero accepted drafts is treated as zero adopted input, or a matching token prefix is used to splice unrelated State and KV checkpoints", + "ResultReady or a commit receipt is treated as DeviceRetired, or selected full-batch intervals are compared with complete-cohort rates under one unlabeled denominator" ], - "refusal_point": "Records missing coverage and high-variance results as INCONCLUSIVE rather than a win; a run without a frozen RunSpec, or whose validator changed after the plan was frozen, is INCONCLUSIVE regardless of metric values; sets no universal numeric gate or noise threshold; withholds causal treatment benefit/null conclusions when required delivery or arm eligibility is unverified, and preserves conflicts between stated protocol, exported code and source reports" + "refusal_point": "Records missing coverage and high-variance results as INCONCLUSIVE rather than a win; a run without a frozen RunSpec, or whose validator changed after the plan was frozen, is INCONCLUSIVE regardless of metric values; sets no universal numeric gate or noise threshold; withholds causal treatment benefit/null conclusions when required delivery or arm eligibility is unverified, and preserves conflicts between stated protocol, exported code and source reports; withholds native state-correctness and cancellation/reuse conclusions when the required same-block oracle, checkpoint identity or matching receipt is missing" }, "recommendations": [ "Freeze one primary objective, timing boundary, quality gates and sample/failure accounting before optimization.", "Choose the required regimes from the deployment use case, report each separately, and record which must improve or must not regress in a versioned decision config.", "Use the replay trace that matches the actual deployment pattern.", "State the concurrency, trace, model, SKU and microbenchmark scope boundaries in any report that cites benchmark results.", - "Trace treatment to its eligible consumer and apply the frozen arm-validity precedence before attributing benefit, saturation or a null." + "Trace treatment to its eligible consumer and apply the frozen arm-validity precedence before attributing benefit, saturation or a null.", + "For hybrid speculative comparisons, freeze the coupled restore point, same-physical-block oracle, zero-draft/zero-adoption controls and complete-cohort versus steady-phase denominators before interpreting a result." ], "limitations": [ "Source-derived candidate; no local efficacy validation.", @@ -274,7 +317,8 @@ "Benchmark feasibility and package-consumption checks do not promote this candidate or replace task quality, repeated A/B and service qualification.", "The pinned NeMo-Speech.cpp ASR reference covers selected source/test definitions and author-reported benchmark boundaries; native model/device, decoder internals, quality, long-audio/load and service behavior remain not_run.", "JustGoal contributes source-only controlled-evaluation and kernel-acceptance guidance; its repaired/retested timings, released-pass counts and protocol differences are author-reported, not locally reproduced or a complete campaign runner.", - "The AutoTuneBench reference is source-only: original operational checkpoint/prompt receipts remain unavailable, source contradictions remain explicit, and no tuning, device/performance or agent-efficacy validation was run." + "The AutoTuneBench reference is source-only: original operational checkpoint/prompt receipts remain unavailable, source contradictions remain explicit, and no tuning, device/performance or agent-efficacy validation was run.", + "The NInfer reference inspects selected Qwen3.5 GDN/MTP source contracts and unexecuted test definitions; its historical author measurements use a different engine revision. Native state/model/device/quality/performance and efficacy remain unvalidated." ], "provenance_readsets": [ "references/task-benchmark-selection.md", @@ -283,11 +327,15 @@ "references/justgoal-evaluation-contracts.md", "references/justgoal-source.json", "references/autotunebench-trial-validity.md", - "references/autotunebench-source.json" + "references/autotunebench-source.json", + "references/ninfer-state-and-completion-contracts.md", + "references/ninfer-source.json" ], "references": [ "../inference-optimization-casebook/cases/admission-drain-occupancy.md", - "../inference-optimization-casebook/cases/admission-drain-occupancy.json" + "../inference-optimization-casebook/cases/admission-drain-occupancy.json", + "../../repositories/deepinfer/budget-speculative-verification/SKILL.md", + "../../repositories/vllm-rlt/audit-recurrent-kv-finalization/SKILL.md" ], "consumption_mode": "packaged", "searchspace_not_applicable": "SKILL.md §§ Select the task and its evaluation contract, Statistical requirements, Shared-machine measurement discipline, and Search integrity and final-state verification design comparable evaluation; hardware identity and allocation isolation are comparison controls, not a hardware-resource mechanism." diff --git a/src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/ninfer-source.json b/src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/ninfer-source.json new file mode 100644 index 000000000..17fbf9227 --- /dev/null +++ b/src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/ninfer-source.json @@ -0,0 +1,529 @@ +{ + "schema_version": 1, + "source": { + "repository": "Neroued/ninfer", + "url": "https://github.com/Neroued/ninfer", + "commit": "1882c6adbc98043fcf89f3589aa5addb3742777a", + "tree": "78ee4a8950a23459e6969fff77e1f72bcb3abd9a", + "repository_id": 1281510950, + "visibility": "public", + "inspected_on": "2026-11-03", + "commit_identity_url": "https://api.github.com/repos/Neroued/ninfer/git/commits/1882c6adbc98043fcf89f3589aa5addb3742777a", + "complete_tree_url": "https://api.github.com/repos/Neroued/ninfer/git/trees/78ee4a8950a23459e6969fff77e1f72bcb3abd9a?recursive=1", + "identity_boundary": "GitHub App commit/tree identity, complete tree and selected archive-transported bytes matched to immutable blob SHA and size; only the listed full/bounded readset is claimed." + }, + "scope": "Qwen3.5 GDN/MTP eager record production, accepted-prefix fold/history repair, coupled checkpoint binding and frontend/Session/local-executor adoption versus retirement contracts; historical serving measurement interpretation.", + "files": [ + { + "path": "LICENSE", + "blob_sha": "261eeb9e9f8b2b4b0d119366dda99c6fd7d35c64", + "bytes": 11357, + "sha256": "c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4", + "role": "rights", + "inspection": "full", + "inspected_line_ranges": [ + [ + 1, + 201 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/LICENSE", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "README.md", + "blob_sha": "aed2772415d6278296b1e07216e36fe50b97ec37", + "bytes": 14437, + "sha256": "4c9d5c49bacc3c13bff598e61f323220e192f2fe424bee931dbcd1416a43da47", + "role": "source identity, deployment and separate model/vendor licenses", + "inspection": "full", + "inspected_line_ranges": [ + [ + 1, + 258 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/README.md", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "docs/maintainer/engine-architecture.md", + "blob_sha": "e2e6ae11ce1de5a1d3b7b9bcba3fbece76786b16", + "bytes": 8806, + "sha256": "35043013053a2d00f7471f229a3c6173cd25a2dd1ef4a15b686f5df5ef4d902b", + "role": "owners, commit/publication/retirement contract", + "inspection": "full", + "inspected_line_ranges": [ + [ + 1, + 133 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/docs/maintainer/engine-architecture.md", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "docs/maintainer/speculative-decoding.md", + "blob_sha": "722b0c74ceb02ed005ebdeb5dbbf99bef39d3f76", + "bytes": 6009, + "sha256": "2a8fbd9a9f1fd60f37af2ee7ecd160465753b8e0045a97fe3f1b919667a360e7", + "role": "accepted inputs, MTP frontiers and represented-input replay contract", + "inspection": "full", + "inspected_line_ranges": [ + [ + 1, + 97 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/docs/maintainer/speculative-decoding.md", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "docs/maintainer/context-management.md", + "blob_sha": "b170ff9ccf09966add64585527f0bc2e0fe9d7e4", + "bytes": 8819, + "sha256": "f0f96c5bf2b3e158fe9b99169600a0d2963fe938a04b59a013d5309227671c97", + "role": "coupled complete State/KV restore and lifetime contract", + "inspection": "full", + "inspected_line_ranges": [ + [ + 1, + 132 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/docs/maintainer/context-management.md", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "docs/performance/methodology.md", + "blob_sha": "76a5690e843a3225887639b47479ea669578635a", + "bytes": 11005, + "sha256": "a512dfbe2c13c1d04a28ad9f7ccd2f20792cdf0283e8cf32b5b68713f03a1cc3", + "role": "historical measurement, full cohort versus selected steady wave", + "inspection": "full", + "inspected_line_ranges": [ + [ + 1, + 175 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/docs/performance/methodology.md", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "docs/performance/qwen3.8-27b.md", + "blob_sha": "0e14ef90314233318f6864464b72a0433ff1f13e", + "bytes": 15116, + "sha256": "5e9076d952fbe9a3f5ffeab121d0531274d4d6a27fd7753f4b29bc58c87357ed", + "role": "historical run identity, source-reported denominators, limits and quality gaps", + "inspection": "bounded", + "inspected_line_ranges": [ + [ + 14, + 46 + ], + [ + 151, + 246 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/docs/performance/qwen3.8-27b.md", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "include/ninfer/ops/gdn_replay.h", + "blob_sha": "02b419a4b33db0ac2ad5a1b011fa03d15a1314e0", + "bytes": 2559, + "sha256": "b54bcb3bd3636e1cc5ade0ef2834792daa6394eec322c54505278327f0dabffe", + "role": "fold row, same physical verify block, zero-row and lifetime contract", + "inspection": "full", + "inspected_line_ranges": [ + [ + 1, + 56 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/include/ninfer/ops/gdn_replay.h", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "src/core/gdn_replay_records.h", + "blob_sha": "17a32e8dfa413ed317e2404c81d674b7f64dc96a", + "bytes": 1761, + "sha256": "3700ee0e2c6483ef4b1eebdd759f52a562f949e00e7b058433f3579fea80b600", + "role": "represented record planes and physical row ownership", + "inspection": "full", + "inspected_line_ranges": [ + [ + 1, + 61 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/core/gdn_replay_records.h", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "src/core/gdn_replay_records.cpp", + "blob_sha": "7305d7f1641e66899c6545300b469d2e921e370d", + "bytes": 6309, + "sha256": "f7846375c0a7aa795e927e7cfe603a0f4e097f7eee0900c2b7a4bce0dd14e8c3", + "role": "record layout and non-owning view geometry", + "inspection": "full", + "inspected_line_ranges": [ + [ + 1, + 142 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/core/gdn_replay_records.cpp", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "src/core/linear_attention_state.h", + "blob_sha": "3c6a4632631afe27483002ce0b29fe36fe18bd28", + "bytes": 1786, + "sha256": "edf7c62ef84855e26ad68c6c29f5cbd896728e79300b2f07f8c027a75746c54a", + "role": "distinct convolution and recurrent fields", + "inspection": "full", + "inspected_line_ranges": [ + [ + 1, + 58 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/core/linear_attention_state.h", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "src/core/linear_attention_state.cpp", + "blob_sha": "4c634106cda183bd2698beee28c3d3eac4868865", + "bytes": 2912, + "sha256": "06cbb20785ade5c6a37cbc830c0220d21614d417851943b20ba31051942b4e19", + "role": "FP32 recurrence and state-image fields", + "inspection": "full", + "inspected_line_ranges": [ + [ + 1, + 64 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/core/linear_attention_state.cpp", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "src/models/qwen3_5/state/state_image.h", + "blob_sha": "42eee24da086819cfce3c730a3dafccbd18bd1ff", + "bytes": 1927, + "sha256": "7dd17e56dc5156d934191c60a15c18cbd93615b9f37ba3b005f01fdcf691733b", + "role": "complete model StateImage fields", + "inspection": "full", + "inspected_line_ranges": [ + [ + 1, + 62 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/models/qwen3_5/state/state_image.h", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "src/models/qwen3_5/state/state_image.cpp", + "blob_sha": "15309a4cb99fb2a2ca52b82c117b1fe026256df0", + "bytes": 1893, + "sha256": "ab67258d80781ed5837663170392d6d2eae0a6bbb1459b717a9ccbc31ddd35b4", + "role": "layout binds recurrence, convolution and continuation hidden", + "inspection": "full", + "inspected_line_ranges": [ + [ + 1, + 51 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/models/qwen3_5/state/state_image.cpp", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "src/models/qwen3_5/execution/gdn.cpp", + "blob_sha": "677ef0440a62ea39953405cb316efb39b6ab3721", + "bytes": 3281, + "sha256": "84599cefcfef7634ee00cf50f0bad49d7b2e6606e523a6c79013ada4159f0976", + "role": "projection record producer wrapper", + "inspection": "full", + "inspected_line_ranges": [ + [ + 1, + 57 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/models/qwen3_5/execution/gdn.cpp", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "src/models/qwen3_5/execution/text.cpp", + "blob_sha": "5f461cf99e26442e55dc8ef7886d0139cf4e4893", + "bytes": 20751, + "sha256": "e3b1874d1440c4288cc4b4708bef36ef45d6c14d80f1465884e4d211c045e8f6", + "role": "complete TextRun::gdn producer branch", + "inspection": "bounded", + "inspected_line_ranges": [ + [ + 184, + 276 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/models/qwen3_5/execution/text.cpp", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "src/models/qwen3_5/program/decode.cpp", + "blob_sha": "3309e753ad857b20b19b9e2ae990b21d827af55b", + "bytes": 10341, + "sha256": "feafd47ce5a5b42956deb20bf0f909759ab93df4229fb638dab2f7644ba1a988", + "role": "complete eager begin_target / advance_target branch", + "inspection": "bounded", + "inspected_line_ranges": [ + [ + 128, + 200 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/models/qwen3_5/program/decode.cpp", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "src/models/qwen3_5/program/commit.cpp", + "blob_sha": "be222c1ac74f01be13f0a5b692833fcd5c2ea067", + "bytes": 9484, + "sha256": "c94b09784238e699caceb88b7d93040a8986427f51458a5581942bb42252dae7", + "role": "fold consumer, MTP repair and DeviceRetired emission", + "inspection": "full", + "inspected_line_ranges": [ + [ + 1, + 179 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/models/qwen3_5/program/commit.cpp", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "src/models/qwen3_5/program/mtp.cpp", + "blob_sha": "202dc7fcd2ad2ade1b7db84d13d8d536b4fbe8ee", + "bytes": 10739, + "sha256": "b781b7a30c9ac863552adcda85613c9f39b604cfa9ea98404ff110ade8eeb85d", + "role": "true target conditioning and accepted-prefix MTP history rebuild", + "inspection": "full", + "inspected_line_ranges": [ + [ + 1, + 209 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/models/qwen3_5/program/mtp.cpp", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "src/models/qwen3_5/session/commit.cpp", + "blob_sha": "39f0f1f40e9a1851b78f815aa58d5291cea94ffc", + "bytes": 20884, + "sha256": "d7235536ed420cd43e049e3a7a37560fa37334bccea39086eeb5fe2aacdb7134", + "role": "generated versus adopted counts, cancel validation, logical adoption and retire publication", + "inspection": "full", + "inspected_line_ranges": [ + [ + 1, + 406 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/models/qwen3_5/session/commit.cpp", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "src/models/qwen3_5/session/binding.cpp", + "blob_sha": "8aa3ac2af61fdf01f1f5330ba3015bfa5f73df22", + "bytes": 21783, + "sha256": "1270f1ff232ce5a3e5c74a05b119f9dc4cb21a54b3addf73c7820907c711d1e4", + "role": "complete start/install/complete binding functions for coupled State/history restore", + "inspection": "bounded", + "inspected_line_ranges": [ + [ + 158, + 399 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/models/qwen3_5/session/binding.cpp", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "src/models/qwen3_5/session/checkpoints.cpp", + "blob_sha": "f1d5e725c9011a62d56cd1ee64b3920f117e58d4", + "bytes": 8991, + "sha256": "4f5dd710735b22d710621079d1739ecdff79fbdd52ada1712eb53af5789bea52", + "role": "checkpoint image and Main/backend history ownership", + "inspection": "full", + "inspected_line_ranges": [ + [ + 1, + 189 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/models/qwen3_5/session/checkpoints.cpp", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "src/runtime/engine/engine_execution.inl", + "blob_sha": "645b8f97090afe2ba94b7c7c1cbaf2a0e0083759", + "bytes": 32366, + "sha256": "96bc53a065bd44dba5279c31318ef6a264fc971209f0e350a97e186e38906b90", + "role": "complete frontend preview → Session commit → published output function", + "inspection": "bounded", + "inspected_line_ranges": [ + [ + 5, + 241 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/runtime/engine/engine_execution.inl", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "src/runtime/executor/contract.h", + "blob_sha": "e0284f68647dfb6ca10449f9de495fadb4eb0cf8", + "bytes": 3729, + "sha256": "e328c86f663f30137a2ea6d8735273fd3e11604c70f9c326b04202d331efd959", + "role": "ticket, commit receipt and DeviceRetired identities", + "inspection": "full", + "inspected_line_ranges": [ + [ + 1, + 144 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/runtime/executor/contract.h", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "src/runtime/executor/local_executor.h", + "blob_sha": "36dbf8fd04518ef384cba9bb0b2120649eba832e", + "bytes": 12366, + "sha256": "2732a40ba85b6cf96f7fb22d023daf28a65adbdb80b41e8c134f46b08902a684", + "role": "receipt returns in Retiring phase, checked retirement releases frame", + "inspection": "full", + "inspected_line_ranges": [ + [ + 1, + 292 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/runtime/executor/local_executor.h", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "src/ops/linear_attention/gated_delta_net/replay/fold.cpp", + "blob_sha": "76f9f13a23254e7d2d5bdd3fe09730237666eeef", + "bytes": 10317, + "sha256": "ac36ca4283dcefb7e28089db2c0e1fc77d239306f0c299e8b7f7f28b3305ed66", + "role": "host zero-row validation and fold dispatch; device metadata boundary", + "inspection": "full", + "inspected_line_ranges": [ + [ + 1, + 177 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/ops/linear_attention/gated_delta_net/replay/fold.cpp", + "license": "Apache-2.0 repository default; no selected-file override identified" + }, + { + "path": "tests/ops/linear_attention/gated_delta_net/test_fold.cpp", + "blob_sha": "d65da514ce709bbf40baab2d17fdcc800ca2833b", + "bytes": 55742, + "sha256": "2092fa936d1495e600f80d0e5d1e09597ebafcfd7abe64cff656a70adbd26dd1", + "role": "inspected same-full-block snapshot/assertion/guard section and registered zero/mixed-prefix cases", + "inspection": "bounded", + "inspected_line_ranges": [ + [ + 330, + 565 + ], + [ + 874, + 965 + ] + ], + "source_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/tests/ops/linear_attention/gated_delta_net/test_fold.cpp", + "license": "Apache-2.0 repository default; no selected-file override identified" + } + ], + "claim_anchors": { + "coupled_checkpoint": [ + "docs/maintainer/context-management.md:27-47,62-77,121-129", + "src/models/qwen3_5/state/state_image.cpp:7-18", + "src/models/qwen3_5/session/binding.cpp:158-399" + ], + "same_physical_block_and_zero_fold": [ + "docs/maintainer/speculative-decoding.md:46-67", + "include/ninfer/ops/gdn_replay.h:19-49", + "src/ops/linear_attention/gated_delta_net/replay/fold.cpp:77-174" + ], + "producer_and_repair": [ + "src/models/qwen3_5/program/decode.cpp:128-200", + "src/models/qwen3_5/execution/text.cpp:184-276", + "src/models/qwen3_5/program/commit.cpp:10-141", + "src/models/qwen3_5/program/mtp.cpp:172-207" + ], + "adoption_and_retirement": [ + "src/models/qwen3_5/session/commit.cpp:203-321,337-404", + "src/runtime/engine/engine_execution.inl:5-241", + "src/runtime/executor/contract.h:44-69,119-133", + "src/runtime/executor/local_executor.h:81-114,116-187" + ], + "unexecuted_test_definitions": [ + "tests/ops/linear_attention/gated_delta_net/test_fold.cpp:330-565,874-965" + ], + "historical_denominators": [ + "docs/performance/methodology.md:44-148", + "docs/performance/qwen3.8-27b.md:14-46,151-246" + ] + }, + "historical_report": { + "reported_execution_revision": "7f6aafedb5f20200def820cfe51ab81c09c20eeb", + "reported_dates": "2026-09-28 to 2026-09-29, Asia/Shanghai", + "reported_hardware_and_stack": "One RTX 5090 32 GiB; driver 617.14; CUDA 13.4; FP8 E4M3 row-256 KV; prefill chunk 1024; prefix reuse disabled; CUDA Graphs and recorded profile-specific artifact/speculation settings.", + "evidence_kind": "author_reported", + "raw_reports": "Local profiles/bench paths named by source report; raw report files were not available in this readset.", + "limits": [ + "Not a measurement of the reference source pin.", + "Selected steady-wave and entire-corpus rates use different time/workload denominators.", + "Output-limit results remain included; request-error absence and token termination do not establish task success.", + "Same-run answer accuracy, prompt compliance and perplexity were not measured." + ] + }, + "rights": { + "selected_software_and_documentation": "Apache-2.0 repository default", + "license_url": "https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/LICENSE", + "license_blob_sha": "261eeb9e9f8b2b4b0d119366dda99c6fd7d35c64", + "license_sha256": "c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4", + "selected_file_findings": "No file-level license override or applicable nested NOTICE identified for the listed readset. README:249-258 explicitly separates model and vendored-dependency licenses.", + "distribution": "Original English procedural prose and immutable public source attribution. InferRSI retains its Apache-2.0 root license. No upstream implementation, model weights, corpus, figure or vendor code is copied.", + "exclusions": [ + "Model-weight and vendor licensing are not audited.", + "This is not a whole-tree license conclusion." + ] + }, + "evidence_boundary": { + "source_contracts_and_code": "inspected_only", + "upstream_test_definitions": "inspected_not_executed", + "native_execution": "not_run", + "model_quality": "not_run", + "performance_reproduction": "not_run", + "agent_efficacy": "not_evaluated", + "future_controls": "Proposed validation observations; no new test implementation, measured tolerance or campaign verdict is supplied.", + "excluded_scope": [ + "KDA/GLM numerical and state generalization", + "Complete DFlash/DFlash2 sampling proofs", + "Current Graph dispatch/native kernel or distributed/Spark qualification", + "Model/backend compatibility, service acceptance or optimization benefit" + ] + } +} diff --git a/src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/ninfer-state-and-completion-contracts.md b/src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/ninfer-state-and-completion-contracts.md new file mode 100644 index 000000000..c575f3440 --- /dev/null +++ b/src/inferrsi/data/skill_library/methods/inference-benchmark-design/references/ninfer-state-and-completion-contracts.md @@ -0,0 +1,90 @@ +# NInfer state-restoration and completion contracts + +Use this reference to freeze a validation plan or audit a supplied hybrid speculative-decoding report. It identifies the state, counts and completion observations a comparison needs. The [source record](ninfer-source.json) fixes NInfer's identity, exact readset and rights. This is candidate knowledge; native correctness, model quality, performance and agent efficacy are unverified. + +The inspected implementation slice is Qwen3.5 GDN/MTP, with an eager producer-to-fold path and the source's public checkpoint and execution contracts. KDA/GLM, DFlash probability math, distributed/Spark behavior and current Graph/kernel qualification are outside this slice. A historical CUDA Graph campaign is a separate author-reported measurement, not evidence that this current implementation was run. + +## Freeze the object being compared + +Record source/build and model/artifact revisions, speculative backend, draft and physical verify widths, sampling/output policy, state and record geometry, dtypes, actual execution path and checkpoint identity. Keep these objects distinct: + +| Object or count | Meaning in the selected source | +|---|---| +| Main frontier `F` | Main has executed inputs `[0,F)`; the known next input can remain an unexecuted anchor | +| MTP history frontier | Stable history reaches `max(F-1,0)` and uses true Main conditioning, not recursive draft hidden | +| Complete checkpoint | Position-related StateImage, appropriate Main/backend histories and their continuation/coverage metadata | +| `accepted_drafts` | Draft positions accepted by target verification; this alone does not state the final adopted prefix | +| Licensed output prefix | Candidate outputs that the frontend may inspect for EOS, stop, budget, grammar failure or cancellation | +| `CommitDecision.accepted_tokens` | Frontend-adopted output extent, which drives the corresponding committed input extent | +| Committed versus accepted draft counters | Adopted draft work and verification acceptance are recorded separately | + +A round adopting `N` outputs commits the corresponding `N` inputs; the last new output becomes the next anchor. Zero accepted drafts can still leave a target-only output to adopt. It must not be used as a zero-state-update control. Cancellation or constraint failure instead requires a terminal decision with zero adopted tokens in the inspected Session contract. + +The [speculative contract](https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/docs/maintainer/speculative-decoding.md#L6-L44) states the alignment. [Session commit](https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/models/qwen3_5/session/commit.cpp#L203-L321) validates final decisions and records the distinct counters. These are source observations, not executed parity results. + +## Establish a coupled restore point + +Identify the StateImage's convolution history, FP32 recurrent state and continuation hidden, plus Main/MTP histories and their frontiers. Freeze their common provenance and valid coverage before selecting a reference run. The inspected binding path prepares both histories, transfers or adopts the state image, restores continuation metadata and reconstructs sampling counts from the committed ledger when required. + +Matching tokens or a prefix digest does not prove that independently recomputed State and KV form one valid checkpoint. Trimming KV alone cannot recover an earlier recurrent state. Preserve model/position/media identity and distinguish logical content from its movable address and from an active consumer's storage lease. A restored replay frontier is also distinct from the first genuinely new unit of work. + +Read the [Context contract](https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/docs/maintainer/context-management.md#L27-L77), selected [StateImage layout](https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/models/qwen3_5/state/state_image.cpp#L7-L18) and [binding functions](https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/models/qwen3_5/session/binding.cpp#L158-L399) for this source boundary. They do not qualify arbitrary checkpoint combinations or another model's state representation. + +## Select the same-block state oracle + +For an authorized native comparison, begin from one initial state and one complete physical verify block. Retain the represented convolution/key/value/gate records actually consumed during that block. Compare a fold of `[0,m)` against the corresponding snapshot in the recurrence trajectory of that **same full block**. Re-running projection at width `m` can change finite-precision evaluation and is a different oracle. + +The [GDN fold contract](https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/include/ninfer/ops/gdn_replay.h#L19-L49) specifies FP32 recurrence and the last three time positions per convolution channel from old history plus the adopted records. It requires identity with the matching same-block snapshot, stable physical record-row identity, disjoint active destinations and preserved operands through their last consumer. This is an intended numerical contract, not a numerical result obtained here. + +The eager producer explicitly selects `ReplayRecords`, records the convolution transition and represented recurrent inputs, and the commit composition folds the chosen prefix. MTP history is then rebuilt from true Main hidden features as needed. Accepting a token does not make recursive draft hidden equivalent to Main hidden, so truncating the draft cache is not a substitute for that repair. Exact producer/consumer anchors are in the source record. + +The inspected [fold test sections](https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/tests/ops/linear_attention/gated_delta_net/test_fold.cpp#L330-L565) define same-block snapshots, byte comparisons and record/source/guard preservation. Their registered cases include zero, mixed and full prefix lengths, with in-place and separate destinations. These are unexecuted source assertions. Future native results must name the actual source build, geometry and executing route; a source read, host declaration or skipped case is not a pass. + +## Freeze negative controls and their observations + +| Control selected before the comparison | Required observation and interpretation | +|---|---| +| No draft position accepted, ordinary target output adopted | Keep `accepted_drafts`, licensed extent and final adopted extent separate; do not expect a zero-state update merely from draft rejection | +| Zero adopted input, including a cancelled terminal row | The GDN fold contract permits no state/record access for a zero row; no adopted prefix is published. Other rows and asynchronous retirement remain separate observations | +| Mixed zero/partial/full rows | Preserve physical row identity; inspect each row's own source, destination and adopted extent instead of compressing the active rows | +| EOS, stop or output budget cuts a licensed prefix | Commit/fold the frontend's final extent and preserve termination reason; rejected or unadopted suffix bytes are not recoverable history | +| Separate source and destination | Check the selected state, source preservation, record immutability and untouched guards; compare the next-use state/logits only under a justified numerical/sampling contract | +| Cancellation while work is in flight | Correlate the actual boundary snapshot and final decision with the same execution ticket; cancellation is not evidence that GPU work or transfers have completed | + +These are proposed validation observations derived from source contracts. They do not introduce test implementations or a universal tolerance. Keep the original native operator/model package and real dependencies when a future check is authorized. Do not replace the source with an extracted formula, mock or a CPU substitute for the required CUDA execution. + +On the first mismatch, retain the initial checkpoint and block identity, first differing row/layer/frontier, represented inputs, final decision and execution identities. Withhold a native state-correctness conclusion while any required oracle, checkpoint identity or receipt is missing. Under stochastic sampling, state-contract checks do not by themselves justify token-for-token equality between runs consuming different random draws. + +## Correlate adoption, publication and retirement + +The selected frontend previews an allowed prefix, Session accepts the model commit and adopts its logical result, and only then does the frontend commit and publish its preview. A commit failure must not become a published token without model state. [The complete runtime function](https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/src/runtime/engine/engine_execution.inl#L5-L241) exposes that order. + +Record the session epoch, sequence/residency generation, frame generation, execution ticket and final command where available. `ResultReady`, `CommitReceipt` and `DeviceRetired` denote different facts. The local executor returns a receipt while the flight is retiring; it releases the frame only after checked retirement for the final command. Session then completes/publishes its pending context transactions and clears cancelled or discarded lanes. The ledger points to each corresponding function. + +A cancel flag, host return or recorded event alone cannot establish storage reuse. Conversely, logical adoption need not wait for host synchronization of every device consumer. Preserve the source's ordering rather than adding a guessed global synchronization to the benchmark. Missing correlation or retirement evidence leaves cancellation/reuse behavior unverified. + +## Interpret historical completion denominators + +The inspected [methodology](https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/docs/performance/methodology.md) and [Qwen3.8 report](https://github.com/Neroued/ninfer/blob/1882c6adbc98043fcf89f3589aa5addb3742777a/docs/performance/qwen3.8-27b.md) describe September 28–29, 2026 author runs at `7f6aafedb5f20200def820cfe51ab81c09c20eeb`. They used one RTX 5090 32 GiB, driver 617.14, CUDA 13.4, FP8 E4M3 row-256 KV, chunk 1,024, disabled prefix reuse and the recorded model/artifact/backend settings. They are not measurements of the source revision used for this reference. Raw report directories are identified in the source page but were not available in this readset. + +| Published quantity | Boundary to preserve | +|---|---| +| Per-request decode phase | Completion tokens minus the first prefill-produced token, divided by that request's decode time | +| Complete corpus decode | Sum of decode tokens divided by the entire 75-request makespan, including prefill, admission waits, transitions and drain | +| Selected steady decode | Committed decode tokens over eligible complete full-batch intervals; excludes ramp-up, prefill, recovery and drain | +| Acceptance | Phase tables average per-request ratios; corpus/wave tables divide summed accepted drafts by summed drafted tokens | +| Termination and task success | Stop token, output limit, request/CUDA error and quality score are separate outcomes | + +For example, the reported NVFP4 C=8 steady wave is 922.4 tok/s, while its full-corpus rate is 490.9 tok/s. Their denominators and workloads differ; taking their ratio does not establish a speedup. Each saturation point is one wave, not a repeated-sample mean. + +All 820 formal requests were reported free of request/CUDA/allocation errors, but this does not establish task completion. The C=1 NVFP4 MTP3 corpus retains 40 stop-token and 35 output-limit outcomes among 75; all five AIME 15 samples reach 65,536 output tokens. The campaign did not measure answer accuracy, prompt compliance or perplexity. Separate model-card scores are not a same-run quality gate. Preserve all samples and report missing quality coverage rather than converting a token budget or natural stop into task success. + +## Route to the existing responsibility + +- For verification-row allocation from acceptance and a cost model, use [budget-speculative-verification](../../../repositories/deepinfer/budget-speculative-verification/SKILL.md). It budgets verification; it does not supply this GDN restore oracle. +- For Ouro LAST_EXITED depth-plane KV propagation and contiguous written-history checks, use [audit-recurrent-kv-finalization](../../../repositories/vllm-rlt/audit-recurrent-kv-finalization/SKILL.md). Its scoped synchronous path excludes speculation/preemption; it is not a replacement for this source's StateImage/frontier contract. +- Use this reference to produce a frozen validation plan or source-bound report interpretation. Model/device parity, quality, repeated performance and service acceptance require separately authorized native evidence and frozen campaign gates. + +## Rights and evidence boundary + +The selected NInfer software/documentation uses Apache-2.0; its exact license identity and selected-file checks are in the source record. This reference uses original procedural prose and attribution, with no copied implementation, model weights, fixture corpus, figure or vendored dependency. NInfer explicitly separates model and vendor licenses from its software license. The selected readset is not a whole-tree or model-weight license audit.