Skip to content

Actions: NVIDIA/TensorRT-LLM

Actions

Bot-Command

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
bot-command.yml
2,500+ workflow runs
2,500+ 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
[https://nvbugs/6341070][fix] Fix scaffolding MajorityVoteController output handling Bot-Command #188968: Issue comment #15925 (comment) created by coderabbitai Bot
8s
8s
[https://nvbugs/6341070][fix] Fix scaffolding MajorityVoteController output handling Bot-Command #188967: Issue comment #15925 (comment) created by KleinBlueC
1s
1s
[None][test] Optimize LLM test startup overhead Bot-Command #188966: Issue comment #15777 (comment) created by tensorrt-cicd
1s
1s
[None][infra] Coerce null CBTS PR diffs to empty string Bot-Command #188965: Issue comment #15874 (comment) created by tensorrt-cicd
1s
1s
[https://nvbugs/6385256][fix] Guard at both layers — add len>0 checks in get_request_num_tokens, return []… Bot-Command #188964: Issue comment #15781 (comment) created by tensorrt-cicd
1s
1s
[None][feat] Support request-scoped capacity-only KV cache compaction Bot-Command #188963: Issue comment #15697 (comment) created by tensorrt-cicd
2s
2s
[https://nvbugs/6385256][fix] Guard at both layers — add len>0 checks in get_request_num_tokens, return []… Bot-Command #188962: Issue comment #15781 (comment) created by reasonsolo
10s
10s
[https://nvbugs/6411931][fix] Append ,fo to the ignore-words-list in pyproject.toml… Bot-Command #188961: Issue comment #15924 (comment) created by coderabbitai Bot
1s
1s
[None][fix] Enable MiniMax M3 piecewise CUDA graphs Bot-Command #188960: Issue comment #15923 (comment) created by tensorrt-cicd
1s
1s
[TRTLLM-13321][feat] Extend rejection sampling to one-model speculative decoding (MTP/DRAFT_TARGET/PARD/DFLASH) Bot-Command #188959: Issue comment #15775 (comment) created by tensorrt-cicd
7s
7s
[None][perf] Fuse Qwen3.5 GDN input projections for attention DP Bot-Command #188958: Issue comment #15884 (comment) created by tensorrt-cicd
1s
1s
[None][fix] Enable MiniMax M3 piecewise CUDA graphs Bot-Command #188957: Issue comment #15923 (comment) created by liji-nv
8s
8s
[TRTLLM-13321][feat] Extend rejection sampling to one-model speculative decoding (MTP/DRAFT_TARGET/PARD/DFLASH) Bot-Command #188956: Issue comment #15775 (comment) created by zhaoyangwang-nvidia
1s
1s
[None][fix] Enable MiniMax M3 piecewise CUDA graphs Bot-Command #188955: Issue comment #15923 (comment) created by coderabbitai Bot
1s
1s
[None][feat] Add Gemma 4 12B Unified (encoder-free multimodal) support Bot-Command #188954: Issue comment #15768 (comment) created by tensorrt-cicd
1s
1s
[None][fix] fix bench-moe benchmark bugs Bot-Command #188953: Issue comment #15397 (comment) created by tensorrt-cicd
1s
1s
[TRTLLM-14029][feat] LTX-2 tile-parallel VAE decode Bot-Command #188952: Issue comment #15753 (comment) created by tensorrt-cicd
1s
1s
[TRTLLM-12557][feat] WideEP FT: add AlltoAll watchdog (1a.3 + 1a.4) Bot-Command #188951: Issue comment #15524 (comment) created by tensorrt-cicd
1s
1s
[None][fix] Support V2 KV cache beam search Bot-Command #188950: Issue comment #15197 (comment) created by tensorrt-cicd
8s
8s
[None][perf] DSV4 o_proj: fuse fp8/UE8M0 quantize into o_a CuTe-DSL e… Bot-Command #188949: Issue comment #15257 (comment) created by tensorrt-cicd
1s
1s
[None][perf] Pre-JIT Mamba SSD HAS_INITSTATES=True kernels during warmup Bot-Command #188948: Issue comment #15876 (comment) created by tensorrt-cicd
7s
7s
[None][feat] VisualGen: Add CuTe DSL FMHA kernels as Source Code JIT Bot-Command #188947: Issue comment #15831 (comment) created by tensorrt-cicd
10s
10s
[TRTLLM-12950][perf] DSv4 follow-up: DeepGEMM and MegaMoE Bot-Command #188946: Issue comment #15632 (comment) created by tensorrt-cicd
8s
8s
[None][fix] Make cache transceiver transport reporting deterministic Bot-Command #188945: Issue comment #15893 (comment) created by tensorrt-cicd
2s
2s
[None][feat] Qwen-Image: load pre-quantized ModelOpt NVFP4/FP8 checkpoints Bot-Command #188944: Issue comment #15470 (comment) created by tensorrt-cicd
1s
1s