Illustration drawn by RepoMasters from the repo’s ID and its language color. Credits

BetaRare

tonbistudio/turboquant-pytorch

From-scratch PyTorch implementation of Google's TurboQuant (ICLR 2026) for LLM KV cache compression. 5x compression at 3-bit with 99.5% attention fidelity.

Owner: tonbistudio (personal account)
Stars: Stars: 1,048
Forks: Forks: 139
Language: Python
License: MIT
Created: 2026
Tier: Beta (700 to 1,999 ★) What this tier means

View on GitHub

More repos by tonbistudio

All 5 repos by tonbistudio

tonbistudio/hermes-multi-agent-workflowStars: 495PythonAlpha
tonbistudio/buzz-skillsStars: 278PythonAlpha
tonbistudio/llm-wikiStars: 269Alpha
tonbistudio/model-traderStars: 125PythonPrototype

Nearby in Python

All Python repos

Supprise0901/TVBox_liveStars: 1,048PythonBeta
volcengine/veScaleStars: 1,048PythonBeta
Soul-AILab/SoulX-SingerStars: 1,048PythonBeta
klen/pylamaStars: 1,048PythonBeta
vimeo/graph-explorerStars: 1,048PythonBeta
Cigaras/IPTV.bundleStars: 1,048PythonBeta