turbo-attn 0.58.0

T1Covered by 1 source · first reported by PyPI / crates.io / RubyGems / Go index / NuGet at 24 Sept, 21:22 UTC

Optimized CUDAgraph-enabled kernels and attention backend for vLLM, SGLang and more based on TurboQuant near-lossless KV cache compression.

releases

Sources

• T1turbo-attn 0.58.0PyPI / crates.io / RubyGems / Go index / NuGet24 Sept, 21:22