Pull requests: NVIDIA/TensorRT-LLM
Pull requests list
[TRTLLM-13694][infra] Add benchmark refresh mappings
#17317
opened Aug 5, 2026 by
Mgluhovskoi
Collaborator
•
Draft
1 task done
[TRTLLM-14867][feat] Auto-tuning for split-K in LoRA grouped GEMM
#17315
opened Aug 5, 2026 by
AlessioNetti
Contributor
Loading…
[TRTLLM-13696][test] Part2.3: Migrate CPU only tests - auto deploy
#17314
opened Aug 5, 2026 by
tongyuantongyu
Member
•
Draft
1 task
[https://nvbugs/6546909][fix] Make the handle's lifetime match its graphs — lazy…
#17313
opened Aug 5, 2026 by
trtllm-agent
Collaborator
Loading…
2 tasks done
[https://nvbugs/6506920][fix] Recover DSpark E2E coverage
#17310
opened Aug 5, 2026 by
longlee0622
Collaborator
Loading…
1 task done
[None][perf] enable zero-copy token passing in KVCacheManagerV2
#17308
opened Aug 5, 2026 by
lowsfer
Member
Loading…
[None][perf] fuse DSpark attention and RMSNorm RoPE
#17307
opened Aug 5, 2026 by
longlee0622
Collaborator
Loading…
[None][fix] Support beam search with C++ KVCacheManagerV2
api-compatible
Accepted LLM API contract change that is backwards-compatible
#17306
opened Aug 5, 2026 by
yizhang-nv
Member
Loading…
1 task done
[#17146][fix] Resolve reasoning mode from the rendered prompt
api-compatible
Accepted LLM API contract change that is backwards-compatible
#17305
opened Aug 5, 2026 by
joerowell
Contributor
Loading…
1 task done
[None][infra] Stop CBTS coverage writes from failing the stage results tar
#17304
opened Aug 5, 2026 by
crazydemo
Collaborator
Loading…
1 task done
[None][test] Adjust perf test cases to avoid OOM and remove outdated test cases
#17303
opened Aug 5, 2026 by
yufeiwu-nv
Collaborator
Loading…
4 tasks done
[https://nvbugs/5945081][fix] un-waive DeepSeek-V3-Lite NVFP4 pp4 CUTLASS test
ci: full pre-merge approved
#17301
opened Aug 5, 2026 by
xxi-nv
Collaborator
Loading…
3 tasks done
[None][test] Allow IB transport in disaggregated tests on Hopper
#17300
opened Aug 5, 2026 by
erictsai-nv
Contributor
Loading…
1 task done
[https://nvbugs/6527301][fix] Do not block ranks with scheduled forward work on disagg gen KV-transfer progress
#17299
opened Aug 5, 2026 by
nv-xtf
Collaborator
Loading…
1 task done
[TRTLLM-13409][fix] bound the perf-sanity harness so a stalled stage fails on its own
#17298
opened Aug 5, 2026 by
JunyiXu-nv
Collaborator
Loading…
[None][test] Add coverage for MsaSparseGqaFmha.forward
#17297
opened Aug 5, 2026 by
StanleySun639
Collaborator
•
Draft
1 task
[None][feat] support NIXL cache transceiver with Ray
#17295
opened Aug 5, 2026 by
chuangz0
Collaborator
Loading…
[https://nvbugs/6507110][chore] Unwaive TestQwen3_30B_A3B::test_nvfp4 dep4_latency_moe_trtllm
#17294
opened Aug 5, 2026 by
sunnyqgg
Collaborator
Loading…
1 task done
[https://nvbugs/6434512][fix] Select Marlin for Qwen3.5 NVFP4 on Hopper
#17293
opened Aug 5, 2026 by
lfr-0531
Collaborator
Loading…
1 task done
[TRTLLM-15017][chore] Unify one-model speculative decoding samplers into SpecSampler
#17292
opened Aug 5, 2026 by
zhaoyangwang-nvidia
Collaborator
Loading…
4 tasks done
[None][test] Unwaive test_ray_disaggregated_serving tp1 and tp2
#17291
opened Aug 5, 2026 by
dominicshanshan
Collaborator
Loading…
1 task done
[https://nvbugs/6482589][fix] Make CuError pickle-safe across process boundaries
#17289
opened Aug 5, 2026 by
lowsfer
Member
Loading…
ProTip!
Add no:assignee to see everything that’s not assigned.