hackimov / turboquant-kvView on GitHub
Open-source PyTorch implementation of Google TurboQuant (ICLR 2026) — extreme KV-cache quantization to ~3 bits with zero accuracy loss. 6x less memory, up to 8x faster inference.
42Mar 26, 2026Updated 3 months ago

Alternatives and similar repositories for turboquant-kv

Users that are interested in turboquant-kv are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.

Sorting:

Are these results useful?