ThinkingCap EXL3 with pinned TensorFold CUDA: measured DGX Spark profile and guarded experimental 24 GB NVIDIA GPU qualification path.
-
Updated
Oct 13, 2026 - Python
#
thinkingcap
Here are 2 public repositories matching this topic...
Qwen3.8-27B and the ThinkingCap fine-tune at 262K context on dual or quad RTX 3090s: Merlin engine v2.5.2 + 11 patches, FP8 KV pool 826,689 (dual) / 1,974,072 (quad), MTP K=4
-
Updated
Oct 13, 2026 - Python
Add this topic to your repo
To associate your repository with the thinkingcap topic, visit your repo's landing page and select "manage topics."