[None][test] Remove RTX-6000 OOM test cases - #12798

Closed

yufeiwu-nv wants to merge 62 commits into

NVIDIA:mainfrom

yufeiwu-nv:fix_L40S

Commits on Mar 31, 2026

Refactor some mistral_small model out of L40S because memory limit
yufeiwu-nv
committed
be9f6b5

Commits on Apr 1, 2026

refact another
yufeiwu-nv
committed
4e768a1
improve robustness
yufeiwu-nv
committed
4e41cba
fix according to AI
yufeiwu-nv
committed
274926b
Merge branch 'main' into fix_L40S
yufeiwu-nv
authored
f8532a0

Commits on Apr 2, 2026

Merge branch 'main' into fix_L40S
yufeiwu-nv
authored
8bb12be
Adjust L40S test list
yufeiwu-nv
committed
d730c6f
Merge branch 'main' into fix_L40S
yufeiwu-nv
authored
0571ad1

Commits on Apr 3, 2026

Merge branch 'main' into fix_L40S
yufeiwu-nv
authored
6fe30f0

Commits on Apr 7, 2026

[None][feat] Support MLA generation in TrtllmGen attention backend (#12606)

yihwang-nv
authored and
yufeiwu-nv
committed
1f9147e
[None][infra] Check in most recent lock file from nightly pipeline

tensorrt-cicd
authored and
yufeiwu-nv
committed
f85e986
[None][infra] No warn_unused_ignores temporarily (#12728)

EmmaQiaoCh
authored and
yufeiwu-nv
committed
bb5a420
[None][feat] Qwen3-Next MTP (#11370)

Izzy Putterman
authored and
yufeiwu-nv
committed
4d24c3d
[None][doc] Blog19 for DWDP. (#12725)

authored and
yufeiwu-nv
committed
6860228
[https://nvbugs/5800591][chore] Unwaive a deepseek MTP test (#12327)

mikeiovine
authored and
yufeiwu-nv
committed
a521574
[None][doc] Add visual generation models to supported models page (#12464)

chang-l
authored and
yufeiwu-nv
committed
f12cc91
[None][infra] Check in most recent lock file from nightly pipeline

tensorrt-cicd
authored and
yufeiwu-nv
committed
057db73
[TRTLLM-11146][feat] Extend python cache transceiver to support nemotron (#12150)

bo-nv
authored and
yufeiwu-nv
committed
f245508
[TRTLLM-11523][feat] Handle different chat template types (#12336)

2ez4bz
authored and
yufeiwu-nv
committed
4e55b2b
[None][infra] Check in most recent lock file from nightly pipeline

tensorrt-cicd
authored and
yufeiwu-nv
committed
e03685b
[#12257][fix] Use the first non-None result returned by hf download workers (#12259)

kev-bi
authored and
yufeiwu-nv
committed
b68e522
[None][feat] Add multi-turn support for trtllm-bench (#12468)

cascade812
authored and
yufeiwu-nv
committed
b78d868
[None][fix] Replace assertions with warnings for unsupported logits/logprobs in speculative sampler (#12547)

yifjiang
authored and
yufeiwu-nv
committed
0841e0d
[None][cleanup] Add supplemental ruff lint for legacy files via ruff-legacy hook (#11469)

venkywonka
authored and
yufeiwu-nv
committed
74fb0e3
[https://nvbugs/5864187][fix] Address H20 Weights Loading OOM for GPTOSS (#11321)

authored and
yufeiwu-nv
committed
4da8d33
[None][feat] Add fused DiT QK Norm + RoPE CUDA kernel for FLUX (#11869)

karljang
authored and
yufeiwu-nv
committed
d14231d
[TRTLLM-9772][feat] Support cache reuse for SSM in KVCacheManagerV2 (#12644)

lowsfer
authored and
yufeiwu-nv
committed
cb375ac
[None][fix] Harmony Parser Delta Grouping + Reuse Report + Better Test Coverage (#12467)

dongfengy
authored and
yufeiwu-nv
committed
f207ee5
[None][feat] MLIR-based auto-generated elementwise fusion for AutoDeploy (#12427)

suyoggupta
authored and
yufeiwu-nv
committed
d7ec1cf
[None][doc] Clean up latest news + blogs, update overview, highlight visual gen (#12753)

laikhtewari
authored and
yufeiwu-nv
committed
5469f54
[None][infra] Check in most recent lock file from nightly pipeline

tensorrt-cicd
authored and
yufeiwu-nv
committed
1f04d8b
[None][feat] Add --custom_tokenizer CLI option to trtllm-bench (#12586)

qiaoxj07
authored and
yufeiwu-nv
committed
e1b1e58
[https://nvbugs/6027560][fix] fix hang issues on DGX_B200-8_GPUs-PyTo… (#12656)

bo-nv
authored and
yufeiwu-nv
committed
1eca31a
[TRTLLM-11597][fix] fix disagg kvcache router for chat API and add disagg benchmark for ai_perf (#12337)

reasonsolo
authored and
yufeiwu-nv
committed
3bded61
[None][fix] Fix Cuda event crash with perf metrics (#12639)

jthomson04
authored and
yufeiwu-nv
committed
1abeb78
[None][fix] Update codes to support nemotron-h corner cases (#12620)

Wanli-Jiang
authored and
yufeiwu-nv
committed
a838539
[None][infra] Waive 10 failed cases for main in post-merge 2636 (#12767)

ZhanruiSunCh
authored and
yufeiwu-nv
committed
be9e004
[https://nvbugs/6018051][fix] Add port conflict retry for disaggregated MP tests (#12618)

reasonsolo
authored and
yufeiwu-nv
committed
e9cddcc
[None][infra] Check in most recent lock file from nightly pipeline

tensorrt-cicd
authored and
yufeiwu-nv
committed
d6f190d
[https://nvbugs/6025177][fix] Fix KV cache issue (#12673)

thorjohnsen
authored and
yufeiwu-nv
committed
8348b0c
[TRTLLMINF-37][feat] Add CI agent failure analysis to L0_MergeRequest… (#12543)

dpitman-nvda
authored and
yufeiwu-nv
committed
29140b8
[None][doc] Update C++ coding guidelines. (#12577)

hnover-nv
authored and
yufeiwu-nv
committed
8f4e03f
[#12324][fix] Fixed wrong token suppressed with ignore_eos in torch sampler (#12358)

MrGeva
authored and
yufeiwu-nv
committed
3a341d0
[https://nvbugs/5849648][fix] Fix GPTOSS Chat Template for Disagg Tests (#12724)

dongfengy
authored and
yufeiwu-nv
committed
c664e4d
[#11094][feat] AutoDeploy transform to fuse silu+mul (#12497)

MrGeva
authored and
yufeiwu-nv
committed
58108cf
[None][infra] Fix source code scanning (#12773)

yuanjingx87
authored and
yufeiwu-nv
committed
6e11a2d
[None][chore] Remove gpu-shell tool from ad-run-agent (#12418)

govind-ramnarayan
authored and
yufeiwu-nv
committed
0e8e1d0
[#9306][cleanup] Remove some fields with redefined defaults (#11671)

2ez4bz
authored and
yufeiwu-nv
committed
103d002
[None][infra] Check in most recent lock file from nightly pipeline

tensorrt-cicd
authored and
yufeiwu-nv
committed
403a425
[None][feat] reuse triton slicing kernel for GDN prefill transpose (#12737)

nv-guomingz
authored and
yufeiwu-nv
committed
b9c1ae8
[None][feat] fix mamba metadata prefill bubble in chunked prefill serving (#12736)

nv-guomingz
authored and
yufeiwu-nv
committed
1654d73
[https://nvbugs/5781731][fix] Unwaive Ray test (#9654)

dominicshanshan
authored and
yufeiwu-nv
committed
0a61d32
[None][fix] Fix outdated argument of readme.md for executorExampleDisaggregated.cpp (#12276)

Fan-Yunfan
authored and
yufeiwu-nv
committed
df8e443
[None][feat] Trtllm-gen FMHA JIT support (#12612)

authored and
yufeiwu-nv
committed
a364e99
[None][infra] Check in most recent lock file from nightly pipeline

tensorrt-cicd
authored and
yufeiwu-nv
committed
e97ebbb
[None][feat] retune causalConv1d fwd dispatch for varlen and short sequences (#12739)

nv-guomingz
authored and
yufeiwu-nv
committed
1fe8b3e
[TRTLLM-9948][infra] Move to use FlexCache in Austin for 5080 nodes (#12615)

EmmaQiaoCh
authored and
yufeiwu-nv
committed
6e8ad57
[TRTLLM-11768][fix] Config updates to enable NVFP4 (#12776)

2ez4bz
authored and
yufeiwu-nv
committed
5405349
[https://nvbugs/6008468][fix] Fix top-K logprobs size for PP (#12623)

pengbowang-nv
authored and
yufeiwu-nv
committed
d1f6dfc
[None][Fix] Remove clone in fp8 quant. (#12687)

Tracin
authored and
yufeiwu-nv
committed
05a3d9e
increase stall handle time
yufeiwu-nv
committed
899be0a
improve to 30 mins
yufeiwu-nv
committed
2451003