Download Latest Version
Release v0.6.7.post1 source code.tar.gz (3.1 MB)
Get an email when there's a new version of FlashInfer
| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| Parent folder | |||
| flashinfer_jit_cache-0.6.6+cu130-cp39-abi3-manylinux_2_28_aarch64.whl | 2026-03-11 | 1.8 GB | |
| flashinfer_jit_cache-0.6.6+cu130-cp39-abi3-manylinux_2_28_x86_64.whl | 2026-03-11 | 1.8 GB | |
| flashinfer_jit_cache-0.6.6+cu129-cp39-abi3-manylinux_2_28_aarch64.whl | 2026-03-11 | 1.6 GB | |
| flashinfer_jit_cache-0.6.6+cu129-cp39-abi3-manylinux_2_28_x86_64.whl | 2026-03-11 | 1.6 GB | |
| flashinfer_jit_cache-0.6.6+cu128-cp39-abi3-manylinux_2_28_aarch64.whl | 2026-03-11 | 1.2 GB | |
| flashinfer_jit_cache-0.6.6+cu128-cp39-abi3-manylinux_2_28_x86_64.whl | 2026-03-11 | 1.2 GB | |
| flashinfer_cubin-0.6.6-py3-none-any.whl | 2026-03-11 | 267.7 MB | |
| flashinfer_python-0.6.6-py3-none-any.whl | 2026-03-11 | 7.8 MB | |
| flashinfer_python-0.6.6.tar.gz | 2026-03-11 | 5.3 MB | |
| README.md | 2026-03-10 | 2.7 kB | |
| Release v0.6.6 source code.tar.gz | 2026-03-10 | 2.9 MB | |
| Release v0.6.6 source code.zip | 2026-03-10 | 3.8 MB | |
| Totals: 12 Items | 9.5 GB | 0 | |
What's Changed
- fix: move ArtifactPath/CheckSumHash imports inside gen_moe_utils_modu… by @dierksen in https://github.com/flashinfer-ai/flashinfer/pull/2681
- Enable sm120f compilation by @kahyunnam in https://github.com/flashinfer-ai/flashinfer/pull/2650
- Ensure -gencode flags are in deterministic order (for ccache) by @benbarsdell in https://github.com/flashinfer-ai/flashinfer/pull/2674
- int16 Block-Scaled State and Stochastic Rounding for SSU (mamba) by @ishovkun in https://github.com/flashinfer-ai/flashinfer/pull/2645
- feat: add pool+indices support to gated_delta_rule_decode_pretranspose (bf16 path) by @kaixih in https://github.com/flashinfer-ai/flashinfer/pull/2619
- chore: replace bare print() with logging across the package by @esmeetu in https://github.com/flashinfer-ai/flashinfer/pull/2648
- fix: reduce smem allocation for tinygemm2 kernel in SM120 by @jimmyzho in https://github.com/flashinfer-ai/flashinfer/pull/2670
- [chore] bench_moe_deepseek.py allows adjusting expert distribution by @rosenrodt in https://github.com/flashinfer-ai/flashinfer/pull/2678
- feat: add support for more MLA head dimensions by @hypdeb in https://github.com/flashinfer-ai/flashinfer/pull/2677
- [fp8_blockwise]Fix int32 overflow in TRTLLM fused MoE activation kernel by @charlotte12l in https://github.com/flashinfer-ai/flashinfer/pull/2642
- Give knam codeowner override for Qwen3.5 (gdn) related directories by @kahyunnam in https://github.com/flashinfer-ai/flashinfer/pull/2680
- HOTFIX: Skip mamba Stochastic Rounding tests on sm_120 by @ishovkun in https://github.com/flashinfer-ai/flashinfer/pull/2699
- chore: Update CODEOWNERS by @flashinfer-bot in https://github.com/flashinfer-ai/flashinfer/pull/2712
- feat: support mxfp4 & mxfp8 entrypoint for blackwell cutedsl dense gemm by @b8zhong in https://github.com/flashinfer-ai/flashinfer/pull/2660
- Undo fix to AutoTuner find_nearest_profile by @danisereb in https://github.com/flashinfer-ai/flashinfer/pull/2697
- Experiment Add @kahyunnam as co-owner for several files by @aleozlx in https://github.com/flashinfer-ai/flashinfer/pull/2713
- chore: Update CODEOWNERS by @flashinfer-bot in https://github.com/flashinfer-ai/flashinfer/pull/2719
- Implement
cutlass_fused_moemxfp8 by @zianglih in https://github.com/flashinfer-ai/flashinfer/pull/2581
New Contributors
- @benbarsdell made their first contribution in https://github.com/flashinfer-ai/flashinfer/pull/2674
- @charlotte12l made their first contribution in https://github.com/flashinfer-ai/flashinfer/pull/2642
- @zianglih made their first contribution in https://github.com/flashinfer-ai/flashinfer/pull/2581
Full Changelog: https://github.com/flashinfer-ai/flashinfer/compare/v0.6.5...v0.6.6
