From 5d310387facb9181176b6f9b51d73ca72b9ce7d2 Mon Sep 17 00:00:00 2001 From: phantom Date: Fri, 6 Nov 2026 06:03:07 +0000 Subject: [PATCH] Release 1.3.1 Manifest version bump to 1.3.1 plus CHANGELOG entry for the merged reasoning-effort extension (PR #11, merged as 3bd8087): - xhigh and max reasoning effort presets for OpenAI-compatible providers - Custom (typed) reasoning effort values in the selector, with whitespace trimming and blank-treated-as-none in the provider - Metadata-only delta on top of main; no code changes in this commit Validation recorded in the CHANGELOG: pinned suite 145 passed / 2 skipped on the PR merge commit; Hassfest, Tests, and HACS Action green on the PR head. --- CHANGELOG.md | 14 ++++++++++++++ custom_components/configurable_llm/manifest.json | 2 +- 2 files changed, 15 insertions(+), 1 deletion(-) diff --git a/CHANGELOG.md b/CHANGELOG.md index a642ce2..589baf7 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -2,6 +2,20 @@ All notable changes to this project are documented here. This project follows [Keep a Changelog](https://keepachangelog.com/en/1.1.0/) conventions and [Semantic Versioning](https://semver.org/spec/v2.0.0.html). +## 1.3.1 + +Community contribution: richer reasoning-effort control for OpenAI-compatible providers. + +### Added + +- **`xhigh` and `max` reasoning effort presets for OpenAI-compatible providers** — the `reasoning_effort` selector gains `xhigh` ("Extra high") and `max` ("Max") options for models that accept them (PR #11 by @AimanFarooq, follow-up to #7). Meta's Model API documents `xhigh` and `max` on some models, and other OpenAI-compatible servers define their own levels. As with the existing options, the provider forwards any value other than `none` verbatim. +- **Custom reasoning effort values** — the selector now accepts a typed custom value, the same way the model picker already does, so provider-specific effort levels outside the standard set (e.g. `ultra`) can be configured directly without waiting for the option to ship in a release. Because custom values are free text, the provider trims stray whitespace and treats a blank value like the `none` default (parameter omitted). Selector labels in `strings.json` and `en.json` cover the new presets and the custom-entry behavior. + +### Validation + +- Full suite on the pinned test tree (HA 2025.8.1, anthropic 0.125.0): **145 passed, 2 skipped** — the 134/2 baseline plus the PR's new forwarding tests (`xhigh`, `max`, custom `ultra`, whitespace trimming) and parametrized config-flow cases +- Hassfest, Tests (Python 3.13), and HACS Action green on PR #11 (merged as `3bd8087`) + ## 1.3.0 Community contribution plus a floor reset: 1.3.0 is the first release that requires Home Assistant 2026.11. diff --git a/custom_components/configurable_llm/manifest.json b/custom_components/configurable_llm/manifest.json index 580f3e3..44a2942 100644 --- a/custom_components/configurable_llm/manifest.json +++ b/custom_components/configurable_llm/manifest.json @@ -21,5 +21,5 @@ "openai>=2.45.0", "voluptuous-openapi>=0.1.0,<0.5" ], - "version": "1.3.0" + "version": "1.3.1" }