atomr-accel-flashattn 0.10.0

FlashAttention v2 + v3 kernel templates for atomr-accel — fp16/bf16/fp8, causal, varlen, ALiBi, sliding window, sink tokens, MQA/GQA, paged KV-cache, and chunked prefill, dispatched through NVRTC + Phase 0.6 cubin cache.
Documentation

atomr-accel-flashattn

There is very little structured metadata to build this page from currently. You should check the main library docs, readme, or Cargo.toml in case the author documented the features in them.

This version has 3 feature flags, 0 of them enabled by default.

default

This feature flag does not enable additional features.

cuda-runtime-tests

This feature flag does not enable additional features.

fp8

This feature flag does not enable additional features.

paged

This feature flag does not enable additional features.