Commit Graph

4585 Commits

Author SHA1 Message Date
Zhiqiang Xie
0eec4cb6cc HiCache, add bench long context plus minor fixs (#9086)
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2025-08-11 16:54:52 -07:00
Faradawn Yang
ff1f68252c [fix] Set Radix tree root node hash to None - Nvidia Dynamo Integration (#9030)
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2025-08-11 14:20:39 -07:00
Zhiqiang Xie
9f78f391ae HiCache Storage: generate hash when inserting new nodes (#9053) 2025-08-11 14:18:59 -07:00
Faraz
f508cd3cb7 TRTLLM-MLA FP8 path (#8638)
Signed-off-by: Faraz Khoubsirat <58580514+farazkh80@users.noreply.github.com>
2025-08-11 14:02:13 -07:00
Xiaoyu Zhang
44e86480e8 fuse allreduce and residual_rmsnorm (#8731) 2025-08-11 13:50:53 -07:00
Lianmin Zheng
8c07fabda7 Update hyperparameter_tuning.md (#9083) 2025-08-11 13:44:11 -07:00
SijiaYang
90f44b74e6 fix: w4afp8 accuracy problem and rebase (#8752)
Signed-off-by: yangsijia.614 <yangsijia.614@bytedance.com>
Co-authored-by: Jinwu <ayrnb@users.noreply.github.com>
2025-08-11 13:41:19 -07:00
Simo Lin
38907fe639 refactor(pd-router): extract common patterns to reduce code duplication (#9081) 2025-08-11 13:32:31 -07:00
Liangsheng Yin
f9afa7dceb Fix docs for clip max new tokens (#9082) 2025-08-11 13:15:21 -07:00
Jimmy
0d9e89ec69 [PD]decode: add CLIP_MAX_NEW_TOKEN for pop_preallocated (#8866) 2025-08-11 13:08:11 -07:00
Hangzhi
3d64fda376 Fix broken Kimi models HuggingFace link (#9080) 2025-08-11 12:15:00 -07:00
633WHU
3bffe11279 Fix chunked prefill size validation for disabled state (#8973)
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2025-08-11 11:05:29 -07:00
HAI
44426e54be Update REVIEWERS (#9063) 2025-08-11 11:04:39 -07:00
ishandhanani
9f24dfefd1 chore(gb200): remove ToT flashinfer installation (#9079) 2025-08-11 11:02:15 -07:00
Yi Zhang
89f1d4f536 update deepep commit to support qwen3-coder (#9066) 2025-08-11 10:42:33 -07:00
Baizhou Zhang
75e6a7cde1 Support radix cache for Lora feature (#7216) 2025-08-11 10:14:11 -07:00
Simo Lin
6f81a710f7 [pd-router] add retry and circuit breakfor for pd router (#9051) 2025-08-11 05:53:26 -07:00
Chang Su
a6452b7188 bugfix: Fix output_ids extraction in detokenizer_manager (#9047) 2025-08-11 03:17:32 -07:00
zhyncs
f4ae50e97c fix: use flashinfer v0.2.11.post1 2025-08-11 02:49:25 -07:00
Yineng Zhang
84cb449eec Revert "chore: upgrade flashinfer 0.2.11 (#9036)" (#9057) 2025-08-11 00:16:39 -07:00
Cheng Wan
f003cd3548 [CI] Fix CI tests (#9050)
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2025-08-10 23:52:05 -07:00
Yineng Zhang
9d834fdcc1 Revert "feat: update flashinfer ar oneshot params (#8687)" (#9054) 2025-08-10 23:24:42 -07:00
Zhiqiang Xie
b32792516a REVIEWERS.md typo fix (#9048) 2025-08-10 22:33:37 -07:00
Simo Lin
067068f271 [router] regular router circuit breaker (#8997) 2025-08-10 21:19:30 -07:00
Lianmin Zheng
6beeff41c5 Update REVIEWERS.md (#9046) 2025-08-10 21:11:14 -07:00
Lianmin Zheng
2e8e7e353b Improve docs and developer guide (#9044) 2025-08-10 21:05:18 -07:00
Lianmin Zheng
2449a0afe2 Refactor the docs (#9031) 2025-08-10 19:49:45 -07:00
Lianmin Zheng
0f229c07f1 Update release-docs.yml (#9037) 2025-08-10 18:52:11 -07:00
Yineng Zhang
dd001a5477 chore: upgrade flashinfer 0.2.11 (#9036) 2025-08-10 17:35:37 -07:00
Lianmin Zheng
4ea9d74a3e Simplify health check (#9034) 2025-08-10 17:35:05 -07:00
Yineng Zhang
dd949ace23 Revert "[1/2][resubmit] sgl-kernel: Fuse routed scaling factor into m… (#9035) 2025-08-10 17:34:54 -07:00
Lianmin Zheng
f2887498f0 Simplify memory pool (#9033) 2025-08-10 17:32:28 -07:00
Stefan He
8ecf6b9d24 Support Flatten Tensor Update Weights to speed up MOE Update Weights by 20% (#8079) 2025-08-10 16:08:59 -07:00
YiXR
0418b9d4ea [Optimization] Update estimated_num_new_pages logic in TokenToKVPoolAllocator (#8794)
Signed-off-by: Xingrui Yi <yixingrui@linux.alibaba.com>
Co-authored-by: Xingrui Yi <yixingrui@linux.alibaba.com>
2025-08-10 16:01:51 -07:00
Lifu Huang
e322a94d1f Reduce CI duration of test_lora_update. (#9024) 2025-08-10 15:34:04 -07:00
Lianmin Zheng
2c7f01bc89 Reorganize CI and test files (#9027) 2025-08-10 12:30:06 -07:00
Lianmin Zheng
b58ae7a2a0 Simplify frontend language (#9029) 2025-08-10 10:59:30 -07:00
Stefan He
6345069f6c [RL] Add test for /abort_request (#7626) 2025-08-10 09:14:19 -07:00
Simo Lin
ce9cf35327 [router] update pyo3 version to 0.25.1 (#9022) 2025-08-10 06:45:51 -07:00
Lifu Huang
f8a173bb50 Improve LoRA Perf by Deprecating FlashInfer and Eliminating Redundant Tensor Ops (#8940) 2025-08-10 01:04:45 -07:00
JiLi
6b847a9a05 Optimize: Cache CUDA device to reduce redundant calls during tensor l… (#8996) 2025-08-10 00:32:57 -07:00
Simo Lin
473400e452 [router] upgrade kube version to latest (#9018) 2025-08-09 22:49:45 -07:00
Simo Lin
dd665f967f [router] upgrade rand to latest version (#9017) 2025-08-09 22:49:30 -07:00
Simo Lin
3817a37d87 [router] upgrade to latest sgl kernel for router ci (#9019) 2025-08-09 21:49:18 -07:00
DarkSharpness
7ba5ad5766 [Fix] Fix flashinfer cpu <-> gpu synchronization (#8340) 2025-08-10 03:11:40 +00:00
DarkSharpness
19bc77f05c [Fix] Fix hicache backend (#8991) 2025-08-09 17:16:25 -07:00
huangtingwei
86497d99f2 fix page first per layer pf2lf kernel (#8915)
Co-authored-by: Zhiqiang Xie <xiezhq@stanford.edu>
2025-08-09 17:16:11 -07:00
cctry
5c31b35db2 [hicache] Optimization for DMA copy (#8245) 2025-08-09 17:16:07 -07:00
Lianmin Zheng
ef48d5547e Fix CI (#9013) 2025-08-09 16:00:10 -07:00
Xiaoyu Zhang
a886564a18 fix flashinfer allreduce fusion import bug (#9007) 2025-08-09 13:47:05 -07:00