Release Checks · Workflow runs · NVIDIA/TensorRT-LLM · GitHub

Actions

All workflows
Workflows
- Release Checks Release Checks
- .github/workflows/test-runner.yml .github/workflows/test-runner.yml
- Auto Assign PR to Author Auto Assign PR to Author
- Auto Full Pre-Merge Approval Label Auto Full Pre-Merge Approval Label
- auto-assign auto-assign
- Blossom-CI Blossom-CI
- Bot-Command Bot-Command
- Clean Up Stale Pull Requests Clean Up Stale Pull Requests
- Close inactive issues Close inactive issues
- CodeQL CodeQL
- CodeRabbit Semantic Conflict Review CodeRabbit Semantic Conflict Review
Management
- Caches
- Deployments

Release Checks

Actions

Loading...
Loading

precommit-check.yml

2,500+ workflow runs

2,500+ workflow runs

[None][feat] Add SiTU support to Rubin FC12 MoE backend Release Checks #80045: Pull request #19468 synchronize by xguannv

2m 31s xguannv:user/xguan/fc12-situ-main

xguannv:user/xguan/fc12-situ-main

2m 31s

[None][infra] Add scheduled CodeRabbit semantic checks Release Checks #80044: Pull request #19621 synchronize by chzblych

2m 28s chzblych:codex/semantic-review-polling

chzblych:codex/semantic-review-polling

2m 28s

[https://nvbugs/6697099][fix] Keep MPI progress alive on the PP sample-state relay thread Release Checks #80043: Pull request #19643 opened by YihuiLu512

2m 16s YihuiLu512:dev-fix-11-J13770-N6385771

YihuiLu512:dev-fix-11-J13770-N6385771

2m 16s

[None][test] Remove 32 closed-bug waive entries for main Release Checks #80042: Pull request #19642 opened by trtllm-agent

2m 7s trtllm-agent:trtllm-ci-report/unwaive-20260927-121401

trtllm-agent:trtllm-ci-report/unwaive-20260927-121401

2m 7s

[None][feat] Add GLM-5.3-Flash (glm5_next) support Release Checks #80041: Pull request #19136 synchronize by ruocheng-nv

2m 14s ruocheng-nv:feat/glm5.3-flash-full-support

ruocheng-nv:feat/glm5.3-flash-full-support

2m 14s

[None][refactor] BREAKING: Remove the Python backend of KVCacheManagerV2 Release Checks #80040: Pull request #19154 synchronize by lowsfer

2m 42s lowsfer:kvcm2-remove-python-backend

lowsfer:kvcm2-remove-python-backend

2m 42s

[None][chore] BREAKING: Test KV cache manager v2 as the default Release Checks #80039: Pull request #17495 synchronize by yizhang-nv

2m 48s yizhang-nv:codex/kvcm-v2-default-true-ci

yizhang-nv:codex/kvcm-v2-default-true-ci

2m 48s

[None][fix] Scope KVCM warmup capacity constraints to DeepSeek V4 Release Checks #80038: Pull request #19213 synchronize by yizhang-nv

2m 39s yizhang-nv:codex/fix-kvcm-v2-init-warmup-budget

yizhang-nv:codex/fix-kvcm-v2-init-warmup-budget

2m 39s

[None][refactor] BREAKING: Remove the Python backend of KVCacheManagerV2 Release Checks #80037: Pull request #19154 synchronize by lowsfer

2m 23s lowsfer:kvcm2-remove-python-backend

lowsfer:kvcm2-remove-python-backend

2m 23s

[None][feat] agent flow: add per-role backend routing & casebook switch Release Checks #80036: Pull request #19522 synchronize by GuanhuaWang2001

2m 0s GuanhuaWang2001:per-role-routing

GuanhuaWang2001:per-role-routing

2m 0s

[TRTLLM-16304][feat] In-tree implementation of staircase Release Checks #80035: Pull request #19056 synchronize by WeiHaocheng

2m 39s WeiHaocheng:feat/staircase-upstream

WeiHaocheng:feat/staircase-upstream

2m 39s

[https://nvbugs/6797544][feat] Unified Dspark KVCache Disagg Release Checks #80034: Pull request #19441 synchronize by allisonlim-nv

2m 53s allisonlim-nv:feat/unified-dspark-kv-cache-disagg

allisonlim-nv:feat/unified-dspark-kv-cache-disagg

2m 53s

[https://nvbugs/6812347][fix] Include the SM107 trtllm-gen context FMHA kernels Release Checks #80033: Pull request #19626 synchronize by zhangcl

1m 39s zhangcl:user/chulianz/rubin-fmha-context-cubins

zhangcl:user/chulianz/rubin-fmha-context-cubins

1m 39s

[None][fix] validate reasoning item content in harmony responses input Release Checks #80032: Pull request #17738 synchronize by lonexreb

2m 34s lonexreb:fix/responses-reasoning-content

lonexreb:fix/responses-reasoning-content

2m 34s

[None][feat] Add GLM-5.3-Flash (glm5_next) support Release Checks #80031: Pull request #19136 synchronize by ruocheng-nv

2m 46s ruocheng-nv:feat/glm5.3-flash-full-support

ruocheng-nv:feat/glm5.3-flash-full-support

2m 46s

[None][feat] Report per-request speculative-decoding acceptance on the response Release Checks #80030: Pull request #19311 synchronize by allisonlim-nv

Action required debermudez:dbermudez/spec-decode-per-request-stats

debermudez:dbermudez/spec-decode-per-request-stats

Action required

[None][fix] Pass max_seq_len to the Qwen-VL attention helper in the MiniCPM-V 4.6 vision encoder Release Checks #80029: Pull request #19641 opened by David-Wu1119

2m 12s David-Wu1119:fix/minicpmv4-6-vision-max-seq-len

David-Wu1119:fix/minicpmv4-6-vision-max-seq-len

2m 12s

[None][fix] add FP16 support to fused QK norm RoPE Release Checks #80028: Pull request #19640 synchronize by atipre

Action required atipre:fix/fused-qk-norm-rope-fp16

atipre:fix/fused-qk-norm-rope-fp16

Action required

[None][refactor] overhaul NVLink one-sided all-to-all Release Checks #80027: Pull request #19610 synchronize by zhangcl

1m 12s zhangcl:user/chulianz/nvlink-one-sided-overhaul

zhangcl:user/chulianz/nvlink-one-sided-overhaul

1m 12s

[None][refactor] overhaul NVLink one-sided all-to-all Release Checks #80026: Pull request #19610 synchronize by zhangcl

1m 31s zhangcl:user/chulianz/nvlink-one-sided-overhaul

zhangcl:user/chulianz/nvlink-one-sided-overhaul

1m 31s

[https://nvbugs/6786712][fix] Re-landed the previously-verified 3-defect payload tree-identically… Release Checks #80025: Pull request #19427 synchronize by reasonsolo

2m 36s tensorrt-cicd:repair-bot-bug6786712

tensorrt-cicd:repair-bot-bug6786712

2m 36s

[None][chore] Remove dead code in model engine Release Checks #80024: Pull request #19513 synchronize by mikeiovine

2m 28s mikeiovine:remove-is-draft-model

mikeiovine:remove-is-draft-model

2m 28s

[None][refactor] Split thop.attention into phased fmha Release Checks #80023: Pull request #17558 synchronize by yihwang-nv

2m 14s yihwang-nv:remove-attention-runner

yihwang-nv:remove-attention-runner

2m 14s

[None][feat] Add GLM-5.3-Flash (glm5_next) support Release Checks #80022: Pull request #19136 synchronize by ruocheng-nv

2m 28s ruocheng-nv:feat/glm5.3-flash-full-support

ruocheng-nv:feat/glm5.3-flash-full-support

2m 28s

[None][test] Skip MoE A2A CFT regression under CUDA forward compatibility Release Checks #80021: Pull request #19639 synchronize by ruocheng-nv

2m 35s ruocheng-nv:fix/moe-a2a-cft-forward-compat-skip

ruocheng-nv:fix/moe-a2a-cft-forward-compat-skip

2m 35s