Skip to content

Actions: flashinfer-ai/flashinfer

Actions

PR Label Cleanup

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
pr-label-cleanup.yml
1,995 workflow runs
1,995 workflow runs

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
feat: add get flashinfer-trace interface .fi_trace PR Label Cleanup #2026: Pull request #2931 synchronize by yyihuang
1s yyihuang:fi_trace
yyihuang:fi_trace
1s
[feat] Trtllm-gen Per-token Nvfp4 MoE PR Label Cleanup #2025: Pull request #3027 synchronize by IwakuraRein
1s IwakuraRein:per-token-fp4
IwakuraRein:per-token-fp4
1s
Add TGV NVFP4 GEMM tactic to mm_fp4 cute-dsl backend (SM100/SM103) PR Label Cleanup #2024: Pull request #3141 synchronize by Sinestro38
3s Sinestro38:tgv-nvfp4-mm-fp4-backend
Sinestro38:tgv-nvfp4-mm-fp4-backend
3s
Add TGV NVFP4 GEMM tactic to mm_fp4 cute-dsl backend (SM100/SM103) PR Label Cleanup #2023: Pull request #3141 synchronize by Sinestro38
Action required Sinestro38:tgv-nvfp4-mm-fp4-backend
Sinestro38:tgv-nvfp4-mm-fp4-backend
Action required
unifying all reduce memory allocation for single-node and multi-node nvlink PR Label Cleanup #2022: Pull request #2955 synchronize by Amir-19
7s Amir-19:nvl_mem_uni
Amir-19:nvl_mem_uni
7s
Fix 10 bugs in BF16 XQA MLA kernel for SM120/SM121 PR Label Cleanup #2021: Pull request #2689 synchronize by blake-snc
10s blake-snc:feat-sm120-bf16-mla
blake-snc:feat-sm120-bf16-mla
10s
[CuTe DSL] Fix FP8 MLA persistent perf regression and ProxyKind cu13 wheel breakage PR Label Cleanup #2020: Pull request #3132 synchronize by saltyminty
1s pgera:pgera/fp8-mma-unroll-fix
pgera:pgera/fp8-mma-unroll-fix
1s
feat: add get flashinfer-trace interface .fi_trace PR Label Cleanup #2019: Pull request #2931 synchronize by yyihuang
1s yyihuang:fi_trace
yyihuang:fi_trace
1s
feat: add get flashinfer-trace interface .fi_trace PR Label Cleanup #2018: Pull request #2931 synchronize by yyihuang
1s yyihuang:fi_trace
yyihuang:fi_trace
1s
[Perf] Add FMHAv2 to flashinfer_benchmark.py and eliminate unnecessary H2D PR Label Cleanup #2017: Pull request #2841 synchronize by jimmyzho
1s jimmyzho:fmhav2-bench
jimmyzho:fmhav2-bench
1s
feat: add get flashinfer-trace interface .fi_trace PR Label Cleanup #2016: Pull request #2931 synchronize by yyihuang
1s yyihuang:fi_trace
yyihuang:fi_trace
1s
feat(MoE): FP8 MoE per-channel quant support (note: need cubin!) PR Label Cleanup #2015: Pull request #2809 synchronize by raayandhar
1s raayandhar:users/rdhar/moe-per-channel-quant
raayandhar:users/rdhar/moe-per-channel-quant
1s
perf(gdn): fix bf16_state T=1 per-call overhead and add pool+padding … PR Label Cleanup #2014: Pull request #3118 synchronize by ameynaik-hub
1s ameynaik-hub:ameyn/bf16_baseline_fix
ameynaik-hub:ameyn/bf16_baseline_fix
1s
feat: add get flashinfer-trace interface .fi_trace PR Label Cleanup #2013: Pull request #2931 synchronize by yyihuang
1s yyihuang:fi_trace
yyihuang:fi_trace
1s
feat(moe-a2a): Update nvlink onesided all-to-all PR Label Cleanup #2012: Pull request #3139 synchronize by trevor-m
1s trevor-m:feat/b802a3818-float32-combine-accumulators
trevor-m:feat/b802a3818-float32-combine-accumulators
1s
feat: add get flashinfer-trace interface .fi_trace PR Label Cleanup #2011: Pull request #2931 synchronize by yyihuang
2s yyihuang:fi_trace
yyihuang:fi_trace
2s
feat: add get flashinfer-trace interface .fi_trace PR Label Cleanup #2010: Pull request #2931 synchronize by yyihuang
1s yyihuang:fi_trace
yyihuang:fi_trace
1s
feat: implement configurable tie_break for filtered topk PR Label Cleanup #2009: Pull request #3095 synchronize by ziang-and
10s zianglih:topk-tie
zianglih:topk-tie
10s
feat: add delimiterless multi-item scoring (MaskMode::kMultiItemScoringV2) PR Label Cleanup #2008: Pull request #2921 synchronize by chanh
Action required chanh:cnguyen/delimiterless-mis-v2
chanh:cnguyen/delimiterless-mis-v2
Action required
feat: Enable FP8 (E4M3/E5M2) in concat_mla_k for optimize long-context prefill performance and refactor type dispatch for BF16/FP16 PR Label Cleanup #2007: Pull request #3129 synchronize by qiching
1s qiching:fix/concat-mla-k-fp8-support
qiching:fix/concat-mla-k-fp8-support
1s
feat: add get flashinfer-trace interface .fi_trace PR Label Cleanup #2006: Pull request #2931 synchronize by yyihuang
1s yyihuang:fi_trace
yyihuang:fi_trace
1s
[Fmha] Add head_dim=512 support for trtllm attention kernels PR Label Cleanup #2005: Pull request #2959 synchronize by yzh119
1s djmmoss:dmoss/trtllm-fmha-head-dim-512
djmmoss:dmoss/trtllm-fmha-head-dim-512
1s
feat: Enable FP8 (E4M3/E5M2) in concat_mla_k for optimize long-context prefill performance and refactor type dispatch for BF16/FP16 PR Label Cleanup #2004: Pull request #3129 synchronize by qiching
1s qiching:fix/concat-mla-k-fp8-support
qiching:fix/concat-mla-k-fp8-support
1s
feat: add delimiterless multi-item scoring (MaskMode::kMultiItemScoringV2) PR Label Cleanup #2003: Pull request #2921 synchronize by chanh
Action required chanh:cnguyen/delimiterless-mis-v2
chanh:cnguyen/delimiterless-mis-v2
Action required
[feat] Trtllm-gen Per-token Nvfp4 MoE PR Label Cleanup #2002: Pull request #3027 synchronize by IwakuraRein
1s IwakuraRein:per-token-fp4
IwakuraRein:per-token-fp4
1s