kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-01-14 06:27:45 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity
859 Commits 25 Branches 59 Tags 2.1 GiB
c67da1fbaa
Commit Graph

859 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Robin Kobus
c67da1fbaa
fix: Eagle decoding in TRT flow (#4229) 
* fix: EagleBuffers lifetime issue

Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>

* refactor: Clean up Eagle kernel parameters

Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>

* fix: Eagle draft tokens init

Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>

* chore: Add check for updated sequence length in TrtGptModelInflightBatching

Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>

* fix: Skip check for beam search

Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>

---------

Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
 2025-05-14 16:10:49 +02:00
Kaiyu Xie
6c45586c51
chore: Remove deprecated Python runtime benchmark (#4171) 
Signed-off-by: Kaiyu Xie <26294424+kaiyux@users.noreply.github.com>
 2025-05-14 18:41:05 +08:00
HuiGao-NV
f4059c6e2e
Add test case for kv memory estimation (#4158) 
* Add test case for kv memory estimation
* Dump running log into file and parse kv cache memory size from file
* Set bigger peak memory size for mixed percision case and test_ptp_quickstart_advanced_eagle3 case
* Revert change to usage of fraction
* use context manager to guard temp files

Signed-off-by: Hui Gao <huig@nvidia.com>
 2025-05-14 18:39:25 +08:00
xinhe-nv
f2bfe2f84f
test: [CI] remove closed bugs (#4207) 
update waive list

Signed-off-by: xinhe-nv <200704525+xinhe-nv@users.noreply.github.com>
 2025-05-14 17:59:05 +08:00
DylanChen-NV
206f82115d
[bug/5247505] fix: CP accuracy on Blackwell (#4188) 
* fix xqa params for cp

Signed-off-by: Dylan Chen <191843203+DylanChen-NV@users.noreply.github.com>

* add test

Signed-off-by: Dylan Chen <191843203+DylanChen-NV@users.noreply.github.com>

* add test

Signed-off-by: Dylan Chen <191843203+DylanChen-NV@users.noreply.github.com>

* try adding B200 multi gpu test

Signed-off-by: Dylan Chen <191843203+DylanChen-NV@users.noreply.github.com>

* add accuracy tests for cp

Signed-off-by: Dylan Chen <191843203+DylanChen-NV@users.noreply.github.com>

---------

Signed-off-by: Dylan Chen <191843203+DylanChen-NV@users.noreply.github.com>
 2025-05-14 17:40:50 +08:00
Anurag Mukkara
b15f57763d
tests: PyTorch multimodal using keyword match (#4215) 
* keyword accuracy check for pytorch multimodal

Signed-off-by: Anurag Mukkara <134339030+amukkara@users.noreply.github.com>

* Change keywords for some prompts

Signed-off-by: Anurag Mukkara <134339030+amukkara@users.noreply.github.com>

* Delete full text answers

Signed-off-by: Anurag Mukkara <134339030+amukkara@users.noreply.github.com>

* Cleanup debug code

Signed-off-by: Anurag Mukkara <134339030+amukkara@users.noreply.github.com>

---------

Signed-off-by: Anurag Mukkara <134339030+amukkara@users.noreply.github.com>
 2025-05-14 17:18:43 +08:00
kanghui0204
5e634dd1bd
feat: Low Precision Allreduce for PCIe based GPU (#3851) 
This PR adds a customized allreduce to TensorRT-LLM. The new allreduce is used for communication on PCIe-based GPUs via low-precision quantization, which can accelerate the PCIe allreduce process.

Signed-off-by: Hui Kang <hkang@nvidia.com>
Co-authored-by: Hui Kang <hkang@nvidia.com>
 2025-05-14 16:45:43 +08:00
Yiqing Yan
a66a02a75a
[Infra] Waive L0 test (#4295) 
Waive L0 test

Signed-off-by: Yiqing Yan <yiqingy@nvidia.com>
 2025-05-14 16:38:33 +08:00
Barry Kang
20b42912ce
[TRTLLM-3330][feat] Support DeepSeek-R1 W4A8 on Hopper (#4123) 
Support DeepSeek-R1 W4A8 on Hopper

Co-authored-by: Barry Kang <43644113+Barry-Delaney@users.noreply.github.com>
Co-authored-by: Jiang Shao <91270701+StudyingShao@users.noreply.github.com>
Signed-off-by: Barry Kang <43644113+Barry-Delaney@users.noreply.github.com>
 2025-05-14 15:48:07 +08:00
Zongfei Jing
bb17649517
test: Add UT for moe trtllmgen (#4258) 
* Add ut for moe trtllmgen

Signed-off-by: Zongfei Jing <20381269+zongfeijing@users.noreply.github.com>

* Update tests/unittest/_torch/modeling/test_modeling_deepseek.py

Co-authored-by: hlu1 <14827759+hlu1@users.noreply.github.com>
Signed-off-by: Zongfei Jing <20381269+zongfeijing@users.noreply.github.com>

---------

Signed-off-by: Zongfei Jing <20381269+zongfeijing@users.noreply.github.com>
Co-authored-by: hlu1 <14827759+hlu1@users.noreply.github.com>
 2025-05-14 15:22:58 +08:00
bhsueh_NV
1a9298bc66
CI: add fp8/fp4 ci on Qwen3-30B-A3B (#4266) 
add fp8/fp4 ci on Qwen3-30B-A3B

Signed-off-by: bhsueh <11360707+byshiue@users.noreply.github.com>
 2025-05-14 14:38:04 +08:00
brb-nv
8280c3d4f2
feat: Support Gemma3-1b-it in Pytorch workflow (#3999) 
Signed-off-by: Balaram Buddharaju <169953907+brb-nv@users.noreply.github.com>
 2025-05-14 14:02:44 +08:00
Yi Zhang
86ae506b9d
[fix] Enable pp tests (#3978) 
Fix misrebase issue

Signed-off-by: Yi Zhang <187001205+yizhang-nv@users.noreply.github.com>
 2025-05-14 10:51:20 +08:00
tburt-nv
58bb34c460
[chore] update CI allowlist 2025-05-13 (#4278) 
update allowlist

Signed-off-by: Tyler Burt <195370667+tburt-nv@users.noreply.github.com>
 2025-05-14 10:41:57 +08:00
Fridah-nv
21dbd163a7
[TRTLLM-5188] fix: [AutoDeploy] unwaive AD build test (#4273) 
* unwaive small build test

Signed-off-by: Ubuntu <201670829+Fridah-nv@users.noreply.github.com>

* unwaive mutigpu/integration tests

Signed-off-by: Ubuntu <201670829+Fridah-nv@users.noreply.github.com>

* fix for torch.compile+flashinfer attention

Signed-off-by: Ubuntu <201670829+Fridah-nv@users.noreply.github.com>

---------

Signed-off-by: Ubuntu <201670829+Fridah-nv@users.noreply.github.com>
 2025-05-14 10:40:12 +08:00
Zhanrui Sun
23b9705bf4
chore: bump version to 0.20.0rc3 (#4261) 
Signed-off-by: ZhanruiSunCh <184402041+ZhanruiSunCh@users.noreply.github.com>
 2025-05-14 10:15:25 +08:00
brb-nv
1ef117688c
test: Validate FP8 and LoRA for Gemma3 (#3670) 
Signed-off-by: Balaram Buddharaju <169953907+brb-nv@users.noreply.github.com>
 2025-05-13 17:28:02 -07:00
Anurag Mukkara
b0a03a289c
fix: Merge PP overlap and non-overlap executor loop (#3878) 
Signed-off-by: Anurag Mukkara <134339030+amukkara@users.noreply.github.com>
 2025-05-14 06:04:36 +08:00
Iman Tabrizian
f408de2d99
Waive disagg kv cache load balancer test (#4276) 
Signed-off-by: Iman Tabrizian <10105175+tabrizian@users.noreply.github.com>
 2025-05-14 06:03:24 +08:00
brb-nv
cd5b3d21a0
feat: Support Mistral Small 3.1 24B VLM in TRT workflow (#4183) 
Signed-off-by: Balaram Buddharaju <169953907+brb-nv@users.noreply.github.com>
 2025-05-14 03:47:22 +08:00
Yiqing Yan
290649b6aa
[Infra] Waive L0 test (#4269) 
Waive L0 test

Signed-off-by: Yiqing Yan <yiqingy@nvidia.com>
 2025-05-13 23:06:13 +08:00
Yiqing Yan
bfa16a63d4
[Infra] Waive L0 test (#4268) 
Waive L0 test

Signed-off-by: Yiqing Yan <yiqingy@nvidia.com>
 2025-05-13 22:43:17 +08:00
Frank
c0c3c7f68c
[TRTLLM-5233][feat]: Add chunking to PyT heuristic for trtllm-bench. (#4133) 
* Add chunking to PyT heuristic.

Signed-off-by: Frank Di Natale <3429989+FrankD412@users.noreply.github.com>

* Cast tokens and batch size to ints.

Signed-off-by: Frank Di Natale <3429989+FrankD412@users.noreply.github.com>

---------

Signed-off-by: Frank Di Natale <3429989+FrankD412@users.noreply.github.com>
 2025-05-13 21:47:06 +08:00
dominicshanshan
44d6adfb68
Waive stress test. (#4262) 
* Waive stress test.

Signed-off-by: Wangshanshan <30051912+dominicshanshan@users.noreply.github.com>

* Apply suggestions from code review

Co-authored-by: Yiqing Yan <yiqingy@nvidia.com>
Signed-off-by: dominicshanshan <30051912+dominicshanshan@users.noreply.github.com>

---------

Signed-off-by: Wangshanshan <30051912+dominicshanshan@users.noreply.github.com>
Signed-off-by: dominicshanshan <30051912+dominicshanshan@users.noreply.github.com>
Co-authored-by: Yiqing Yan <yiqingy@nvidia.com>
 2025-05-13 21:01:57 +08:00
Yukun He
cbca6505ff
[nvbugs/5268808][fix] Fix the list-out-of-range access issue of AllReduce workspace on multi-node. (#4159) 
This issue is found for tp=ep=8 on the multi-node machine due to the inconsistent PP sizes.
* Reform the workspace allocation implementation to avoid the list-out-of-range issues.
* Disable min_latency_mode under the multi-node scenario to avoid the illegal memory access issue.

Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com>
 2025-05-13 17:17:25 +08:00
Enwei Zhu
8f68d56cc1
[https://nvbugs/5220763] [test] Unwaive Mixtral FP8 TP2 test (#4252) 
unwaive

Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>
 2025-05-13 15:55:33 +08:00
Yiqing Yan
fda8b0277a
[Infra][TRTLLM-4374] Upgrade TRT 10.10.0 GA, CUDA 12.9 GA and DLFW 25.04 (#4049) 
* [TRTLLM-4374] Upgrade TRT 10.10.0 GA, CUDA 12.9 GA and DLFW 25.04

Signed-off-by: Yiqing Yan <yiqingy@nvidia.com>

* fix review

Signed-off-by: Yiqing Yan <yiqingy@nvidia.com>

* update images

Signed-off-by: Yiqing Yan <yiqingy@nvidia.com>

* Update jenkins/L0_Test.groovy

Co-authored-by: Yanchao Lu <yanchaol@nvidia.com>
Signed-off-by: Yiqing Yan <yiqingy@nvidia.com>

* update image name

Signed-off-by: Yiqing Yan <yiqingy@nvidia.com>

---------

Signed-off-by: Yiqing Yan <yiqingy@nvidia.com>
Co-authored-by: Yanchao Lu <yanchaol@nvidia.com>
 2025-05-13 14:59:12 +08:00
Perkz Zheng
e8d7834c50
fix: [https://nvbugspro.nvidia.com/bug/5238626] illegal memory address when running llama 4 with cuda graph enabled (#4101) 
Signed-off-by: Perkz Zheng <67892460+PerkzZheng@users.noreply.github.com>
 2025-05-13 14:58:54 +08:00
v-shobhit
1770dd96d8
Fix Pipeline Parallelism in Llama4 (#4106) 
Signed-off-by: Shobhit Verma <shobhitv@nvidia.com>
 2025-05-12 22:54:37 -07:00
nvpohanh
13c8e5a8a8
feat: Prefetch safetensors files before loading them (#4140) 
Prefetching safetensors files so that they are stored in the system file
cache. This significantly speeds up the model weight loading for the
very first run after entering the docker container.

This is beneficial because model weight loading is done layer-by-layer,
which means reading from the safetensors chunk-by-chunk, and that cannot
utilize the internet bandwidth very well, assuming that these files are
stored in some network drives. Instead, loading the whole files in bulk
can achieve higher internet bandwidth utilization.

When running with world_size>1, all ranks collaboratedly prefetch these
files.

In theory, we should add heuristics to decide whether to prefetch the
files or not, but that is beyond the scope of this commit.

For example, when the CPU memory is small, doing prefetching may result
in file cache thrashing, resulting in slower weight loading time.

Signed-off-by: Po-Han Huang <pohanh@nvidia.com>
 2025-05-13 13:35:30 +08:00
bhsueh_NV
24be357964
doc: update qwen3 document (#4246) 
* update qwen3 document

Signed-off-by: bhsueh <11360707+byshiue@users.noreply.github.com>

* remove wrong note

Signed-off-by: bhsueh <11360707+byshiue@users.noreply.github.com>

---------

Signed-off-by: bhsueh <11360707+byshiue@users.noreply.github.com>
 2025-05-13 11:05:47 +08:00
ruodil
d555fe2530
test: fix for perf test script issue (#4230) 
fix for perf test script issue

Signed-off-by: Ruodi <200874449+ruodil@users.noreply.github.com>
Co-authored-by: Larry <197874197+LarryXFly@users.noreply.github.com>
 2025-05-13 10:29:20 +08:00
xinhe-nv
0cebc16139
test: [CI] Add failed cases into waives.txt (#4205) 
waive tests

Signed-off-by: xinhe-nv <200704525+xinhe-nv@users.noreply.github.com>
 2025-05-13 10:22:42 +08:00
xinhe-nv
7ebae4dcaa
test: [CI] Add failed cases into waives.txt (#4203) 
* update waive list

Signed-off-by: xinhe-nv <200704525+xinhe-nv@users.noreply.github.com>

* update waives

Signed-off-by: xinhe-nv <200704525+xinhe-nv@users.noreply.github.com>

---------

Signed-off-by: xinhe-nv <200704525+xinhe-nv@users.noreply.github.com>
 2025-05-13 10:08:02 +08:00
pcastonguay
9643be5f20
[TRTLLM-5050][feat] Enable per-request stats with PyT backend (#4156) 
* feat: Add per-request stats support with PyT backend

Signed-off-by: Patrice Castonguay <55748270+pcastonguay@users.noreply.github.com>

* Adding unit test

Signed-off-by: Patrice Castonguay <55748270+pcastonguay@users.noreply.github.com>

* Fixing stats unit test

Signed-off-by: Patrice Castonguay <55748270+pcastonguay@users.noreply.github.com>

* Fixing test with overlap

Signed-off-by: Patrice Castonguay <55748270+pcastonguay@users.noreply.github.com>

---------

Signed-off-by: Patrice Castonguay <55748270+pcastonguay@users.noreply.github.com>
 2025-05-12 21:35:15 -04:00
Simeng Liu
286a789549
feat: Add heuristic for GroupRMSNorm kernel selection. (#4047) 
* feat: Add heuristic for GroupRMSNorm kernel selection.

Implements a logistic regression model to dynamically select between:
- GroupRMSNormBaseKernel: Allocates warps proportional to sum of dimensions
  (better SM occupancy in most cases)
- GroupRMSNormLargeBatch: Allocates warps proportional to max dimension
  (better block scheduling in large batch scenarios)

Selection heuristic considers batch size, allocated warps, and scheduling
efficiency on the current GPU architecture. Models for Compute Capability
9.x and 10.x are trained base on nsys kernel runtime data.
The default kernel selection is the base kernel.

The python operator group_rms_norm will use the heuristic by default.
User can pick to use the base or large batch kernels as well.

Signed-off-by: Simeng Liu <simengl@nvidia.com>

* Address the comments.

Signed-off-by: Simeng Liu <simengl@nvidia.com>

---------

Signed-off-by: Simeng Liu <simengl@nvidia.com>
 2025-05-13 08:52:53 +08:00
Erin
4becf32360
fix: reshape token_ids for lp in torch backend (#4239) 
reshape token_ids

Signed-off-by: Erin Ho <14718778+hchings@users.noreply.github.com>
 2025-05-13 08:43:47 +08:00
Enwei Zhu
035d915fea
[TRTLLM-5081] [test] Align parametrize_with_ids to the pytest behavior (#4090) 
* fix

Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>

* fix

Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>

* normalize mtp_nextn

Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>

* fix

Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>

* fix

Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>

* update test_durations

Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>

---------

Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>
 2025-05-13 07:41:51 +08:00
wili
eba3623a54
Feat: Variable-Beam-Width-Search (VBWS) part4 (#3979) 
* feat/vbws-part4-v1.8: rebase

Signed-off-by: wili-65535 <wili-65535@users.noreply.github.com>

* feat/vbws-part4-v1.9: fix incorrect output when using short output length

Signed-off-by: wili-65535 <wili-65535@users.noreply.github.com>

* v1.9.1: remove useless variables

Signed-off-by: wili-65535 <wili-65535@users.noreply.github.com>

* v1.9.2:fix incorrect output when using short output length

Signed-off-by: wili-65535 <wili-65535@users.noreply.github.com>

* v1.9.3: rebase

Signed-off-by: wili-65535 <wili-65535@users.noreply.github.com>

* v1.9.4: rebase

Signed-off-by: wili-65535 <wili-65535@users.noreply.github.com>

* v1.9.5: remove API change

Signed-off-by: wili-65535 <wili-65535@users.noreply.github.com>

---------

Signed-off-by: wili-65535 <wili-65535@users.noreply.github.com>
Co-authored-by: wili-65535 <wili-65535@users.noreply.github.com>
 2025-05-12 22:32:29 +02:00
yuxianq
a4c3359513
fix: Reset planned states to avoid memory leak in TrtllmAttentionWrapper (#4227) 
Signed-off-by: Yuxian Qiu <142763828+yuxianq@users.noreply.github.com>
 2025-05-12 23:25:54 +08:00
Fridah-nv
3dbb087292
[TRTLLM-5188] fix: [AutoDeploy] update output shape of prepare_fused_mha_metadata_fake (#4199) 
* update output shape of fake kernel prepare_fused_mha_metadata_fake

Signed-off-by: Frida Hou <201670829+Fridah-nv@users.noreply.github.com>

* minor

Signed-off-by: Frida Hou <201670829+Fridah-nv@users.noreply.github.com>

---------

Signed-off-by: Frida Hou <201670829+Fridah-nv@users.noreply.github.com>
 2025-05-12 11:11:40 -04:00
Robin Kobus
b1bee9c394
Revert "Add initial list of CODEOWNERS (#4105)" (#4234) 
This reverts commit aa7300e040.

Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
 2025-05-12 16:53:49 +02:00
Yiteng Niu
31a2e2d08d
doc: update switcher.json config (#4220) 2025-05-12 20:40:55 +08:00
Enwei Zhu
c31ca1688c
[https://nvbugs/5214229] [fix] Unwaive lm_head quantization case (#4222) 
unwaive

Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>
 2025-05-12 20:23:06 +08:00
yuxianq
b35f9a67f9
refactor: Allow models to override apply_qk_norm. (#4078) 
Signed-off-by: Yuxian Qiu <142763828+yuxianq@users.noreply.github.com>
 2025-05-12 19:38:24 +08:00
Zheng Duan
c9e2a963e0
feat: add kv cache aware router (#3831) 
* kv cache aware router

Signed-off-by: Zheng Duan <200704041+zhengd-nv@users.noreply.github.com>

* add tests

Signed-off-by: Zheng Duan <200704041+zhengd-nv@users.noreply.github.com>

* router config

Signed-off-by: Zheng Duan <200704041+zhengd-nv@users.noreply.github.com>

* eviction test

Signed-off-by: Zheng Duan <200704041+zhengd-nv@users.noreply.github.com>

add test

Signed-off-by: Zheng Duan <200704041+zhengd-nv@users.noreply.github.com>

* eviction detect in worker test

Signed-off-by: Zheng Duan <200704041+zhengd-nv@users.noreply.github.com>

* move worker tests to single gpu

Signed-off-by: Zheng Duan <200704041+zhengd-nv@users.noreply.github.com>

* reduce memory fraction

Signed-off-by: Zheng Duan <200704041+zhengd-nv@users.noreply.github.com>

* fix partial block

Signed-off-by: Zheng Duan <200704041+zhengd-nv@users.noreply.github.com>

---------

Signed-off-by: Zheng Duan <200704041+zhengd-nv@users.noreply.github.com>
 2025-05-12 07:23:57 -04:00
Yixin Dong
c90ebadd84
feat: Support the Structural Tag in guided decoding (#4066) 
* finish

Signed-off-by: Ubospica <ubospica@gmail.com>

* update

Signed-off-by: Ubospica <ubospica@gmail.com>

* update

Signed-off-by: Ubospica <ubospica@gmail.com>

* fix

Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>

* exc overlap scheduler

Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>

* add test

Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>

* fix api ref

Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>

---------

Signed-off-by: Ubospica <ubospica@gmail.com>
Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>
Co-authored-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>
 2025-05-12 17:24:50 +08:00
Yechan Kim
3e9bda3a09
[feat] Support HyperCLOVAX-SEED-Text language part (#3902) 
* feat: support HyperCLOVAX-SEED-Text language part

Signed-off-by: yechank <161688079+yechank-nvidia@users.noreply.github.com>

* add Pytorch flow and remove test file

Signed-off-by: yechank <161688079+yechank-nvidia@users.noreply.github.com>

* revert summarize

Signed-off-by: yechank <161688079+yechank-nvidia@users.noreply.github.com>

* fix summarize

Signed-off-by: yechank <161688079+yechank-nvidia@users.noreply.github.com>

* remove from pytorch example

Signed-off-by: yechank <161688079+yechank-nvidia@users.noreply.github.com>

---------

Signed-off-by: yechank <161688079+yechank-nvidia@users.noreply.github.com>
Co-authored-by: QI JUN <22017000+QiJune@users.noreply.github.com>
 2025-05-12 16:05:14 +08:00
Martin Marciniszyn Mehringer
33977dbd42
infra: [TRTLLM-325] Prepare for NGC release - multiplatform build (#4191) 
* infra: [TRTLLM-325] Prepare for NGC release - prepare multiplatform build

Signed-off-by: Martin Marciniszyn Mehringer <11665257+MartinMarciniszyn@users.noreply.github.com>
 2025-05-12 00:38:45 -07:00
Perkz Zheng
3f29d2f006
Feat: support exporting softmax statistics and update the kernel-selection heuristic (#4155) 
* update cubins

Signed-off-by: Perkz Zheng <67892460+PerkzZheng@users.noreply.github.com>

* support exporting softmax statistics and update the kernel-selection heuristic

Signed-off-by: Perkz Zheng <67892460+PerkzZheng@users.noreply.github.com>

---------

Signed-off-by: Perkz Zheng <67892460+PerkzZheng@users.noreply.github.com>
 2025-05-12 15:31:46 +08:00