AlphaUncommon
sudoingX/qwen38-mtp
One llama.cpp flag unlocks +33-39% decode speed for Qwen3.8-27B on consumer GPUs. The MTP head already ships inside your GGUF. Recipe, paired benchmarks, probe tool.
- Owner
- sudoingX (personal account)
- Stars
- Stars: 365
- Forks
- Forks: 102
- Language
- Python
- License
- Apache-2.0
- Created
- 2026
- Tier
- Alpha (250 to 699 ★) What this tier means
More repos by sudoingX
All 3 repos by sudoingX- sudoingX/local-ai-on-your-gaming-gpuStars: 169PythonPrototype
- sudoingX/octopus-invadersStars: 106JavaScriptPrototype
Nearby in Python
All Python repos- emwstudio/VoxEMWStars: 365PythonAlpha
- chemark/ai-agent-bookStars: 365PythonAlpha
- chiba233/atm10-zh-cnStars: 365PythonAlpha
- NeuralEnsemble/python-neoStars: 365PythonAlpha
- benoitc/gafferStars: 365PythonAlpha
- nigma/django-easy-pdfStars: 365PythonAlpha