PrototypeCommon
jayleaton/glm53-tensorfold-spark
GLM-5.3-Flash (abliterated EXL3) on 2x NVIDIA DGX Spark with the TensorFold engine: 1.8x faster decode than vLLM, 4x256k concurrent threads, byte-exact speculative decoding. Work in progress.
- Owner
- jayleaton (personal account)
- Stars
- Stars: 120
- Forks
- Forks: 11
- Language
- Python
- License
- Apache-2.0
- Created
- 2026
- Tier
- Prototype (100 to 249 ★) What this tier means
More repos by jayleaton
All 2 repos by jayleaton- jayleaton/deepseek-v41-tensorfold-sparkStars: 124ZigPrototype
Nearby in Python
All Python repos- Addition2000/mcd-diet-guardStars: 120PythonPrototype
- xiaozhangup/Oh-My-McDonaldStars: 120PythonPrototype
- Aevonix/FlashAlignStars: 120PythonPrototype
- henriklied/django-twitter-oauthStars: 120PythonPrototype
- rtyler/SpawningStars: 120PythonPrototype
- superfell/BeatboxStars: 120PythonPrototype