[pull] main from NVIDIA:main#278

Merged

pull[bot] merged 8 commits intoyingguo-trt:mainfrom

Mar 12, 2026

Commits on Mar 12, 2026

[None][chore] Add failed cases into waives.txt (#12093)
xinhe-nv
authored
[TRTLLM-11092][feat] add support for visual gen FA4 attention backend (#11697)
o-stoner
authored
[https://nvbugs/5955173][fix] Add abort method for GenerationResultBase (#11970)
JunyiXu-nv
authored
[None][test] Add speculative decoding test with exclude_input_in_output=true (#12080)
StanleySun639
authored
[None][feat] Add shared expert LoRA support for MoE models in PyTorch backend (#11760)
achartier
authored
[https://nvbugs/5846166][bug] Fix Disagg Perf Test's MPI Issue and Port Conflict (#12020)
chenfeiz0326
authored
[TRTLLM-10244][doc] Add deployment guide for Nemotron 3 Super (#12129)
nv-guomingz
authored
[None][fix] Narrow bare except clause and use identity check for None (#12041)

edenfunf
and
pengbowang-nv
authored