git
ask
hub
Privacy
Terms
Sign in with GitHub
thetom/turboquant_plus
↗
Stars ·
6,991
Language ·
Python
License ·
Apache-2.0
Simplify and Visualize This Repo
Scan the safety of this repo
Self Host this repo
Check out more work of Developer: thetom
Ask anything about this repo to start.
How do I configure TurboQuant+ to run a 70B model with turbo4 cache compression on an NVIDIA RTX 3080?
Show me how to enable first-and-last-layer protection in TurboQuant+ to recover output quality when using turbo2 compression.
Build a TurboQuant+ setup on Apple Silicon Mac to run a large language model within available unified memory.
Compare turbo2 vs turbo4 cache formats in TurboQuant+ by running the same prompt and measuring the quality difference.
Install TurboQuant+ prebuilt binaries on Windows and run a model without needing any build tools or CUDA setup.
Full explanation on explaingit →
📎
Send
By chatting or signing in you agree to the
Terms
and chat-message logging (revocable in
History
).