|
7f67965b4d
|
[BugFix] NCCL deadlock in HiCache writing_check by making all_reduce unconditional (#26923)
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Zhiqiang Xie <xiezhq@stanford.edu>
Co-authored-by: Zhangheng <hzh0425@apache.org>
|
2026-06-22 01:47:44 +08:00 |
|
|
iridiumine
|
5351800700
|
[Bugfix] Fix MTP acceptance regression on plan stream by moving int64 cast before plan stream context (#28410)
|
2026-06-22 01:26:58 +08:00 |
|
|
Mohammad Miadh Angkad
|
643ee748c6
|
[PP] Pass DSA topk through PP warmup proxy buffers (#28785)
|
2026-06-21 23:55:39 +08:00 |
|
|
Lianmin Zheng
|
7942d546d1
|
Revert "[Spec] Split init_backends; account draft weights in --mem-fraction-static" (#28841)
|
2026-06-21 07:52:35 -07:00 |
|
|
Mick
|
320b231ea6
|
[diffusion] chore: bound DiffGenerator local cleanup (#28833)
|
2026-06-21 22:23:49 +08:00 |
|
|
Mick
|
a7f31a6e1b
|
[diffusion] fix: fix SANA-WM CFG-parallel tensor devices (#28835)
|
2026-06-21 22:01:00 +08:00 |
|
|
Mick
|
a51d56d948
|
CI: Pin flash-attn-4 for diffusion CI consistency (#28838)
|
2026-06-21 21:13:41 +08:00 |
|
|
Lianmin Zheng
|
3975ea5ac7
|
Fix H20 torch import reinstall fallback (#28818)
|
2026-06-21 04:50:46 -07:00 |
|
|
cctry
|
9691a29fe0
|
[Spec] Split init_backends; account draft weights in --mem-fraction-static (#28683)
|
2026-06-21 01:22:26 -07:00 |
|
|
cctry
|
6d4ca9bc54
|
Cap SWA pool sizing with chunk cache (#28755)
|
2026-06-21 01:06:59 -07:00 |
|
|
Lianmin Zheng
|
c9488241e9
|
[Refactor] Auto-derive CLI args from dataclass fields to eliminate duplication (#28814)
|
2026-06-21 00:51:08 -07:00 |
|
|
Jairo David Campaña RoseroandXinyuan Tong
|
b4dda8b3ce
|
fix(anthropic): handle mid-conversation system messages (#26773)
Co-authored-by: Xinyuan Tong <115166877+JustinTong0323@users.noreply.github.com>
|
2026-06-21 04:34:47 +00:00 |
|
|
Mick
|
c65f4ea692
|
[diffusion] fix: validate openai sampling dimensions (#28791)
|
2026-06-21 09:44:38 +08:00 |
|
|
Mick
|
6a16573a7f
|
[diffusion] fix: fix Qwen-Image-Layered string image paths (#28790)
|
2026-06-21 09:43:56 +08:00 |
|
|
Lianmin Zheng
|
d331fdd2ba
|
Add project rule: prefer msgspec.Struct over dataclasses (#28816)
|
2026-06-20 18:43:16 -07:00 |
|
|
Lianmin Zheng
|
54b9b9d0c9
|
Remove threading atexit monkey patch (#28812)
|
2026-06-20 18:30:50 -07:00 |
|
|
karverma-amd
|
2552b860a3
|
[AMD][bugfix] Place TBO cuda-graph num_token_non_padded buffer on model devices (#28337)
|
2026-06-20 18:06:22 -07:00 |
|
|
pure water
|
5b3eeaf504
|
[Fix] MM pool GPU alloc with base_gpu_id (#23377)
|
2026-06-21 08:51:06 +08:00 |
|
|
Polisetty V R K Jyothendra Varma
|
fbbf559de2
|
fix bench_one_batch by extending array with array not list (#28732)
Signed-off-by: P V R K Jyothendra Varma <polisetty.v.r.k.jyothendra.varma@intel.com>
|
2026-06-21 08:28:41 +08:00 |
|
|
Lianmin Zheng
|
8a3d6c3403
|
Sort pyproject dependency lists (#28811)
|
2026-06-20 17:24:57 -07:00 |
|
|
|
f42ec350b4
|
[mtp] add rejection sampling for speculative decoding (#26312)
Co-authored-by: lyc508653 <lyc508653@alibaba-inc.com>
Co-authored-by: Qiaolin-Yu <liin1211@outlook.com>
Co-authored-by: Huiqiang Jiang <30883354+iofu728@users.noreply.github.com>
Co-authored-by: Yi Zhang <25844240+yizhang2077@users.noreply.github.com>
Co-authored-by: Yizhong Cao <114661107+cao1zhg@users.noreply.github.com>
|
2026-06-20 15:10:42 -07:00 |
|
|
Lianmin Zheng
|
95fb1ef697
|
[CI] Remove deprecated test/srt legacy CI setup (#28810)
|
2026-06-20 15:09:33 -07:00 |
|
|
Lianmin Zheng
|
fe428dd845
|
Clean up startup log noise (#28807)
|
2026-06-20 15:02:52 -07:00 |
|
|
Rita BrugarolasandClaude Opus 4.6
|
d6d06cdc17
|
[AMD] Fix no-op dtype cast in _topk_ids_logical_to_physical_dynamic on HIP (#28074)
Signed-off-by: Rita Brugarolas Brufau <rita.brugarolasbrufau@amd.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
|
2026-06-20 11:07:22 -07:00 |
|
|
shuwenn
|
ff1fc1fbdf
|
[mem_cache][5/N] refactor: extract host KV cache base layer into pool_host package (#27273)
|
2026-06-20 20:44:08 +08:00 |
|
|
Mick
|
1109acc24b
|
[diffusion] optimize: shard hunyuan text tokens under sp (#28319)
|
2026-06-20 18:21:38 +08:00 |
|
|
Mick
|
a38eba0f1a
|
[diffusion] CI: remove flaky layerwise offload diffusion case (#28778)
|
2026-06-20 17:27:27 +08:00 |
|
|
kkandwunhuang
|
47cad39f34
|
[AMD] Optimize o_proj gemm and attn output rope performance (#28722)
Co-authored-by: wunhuang <wunhuang@amd.com>
|
2026-06-20 02:11:01 -07:00 |
|
|
Zhiyao JiangandXinyu Jiang
|
1115373668
|
[AMD] Fix garbled unquantized Qwen3-30B-A3B output on ROCm/aiter where the aiter CK fused-MoE falls back to Triton with pre-shuffled weights (#28244)
Co-authored-by: Xinyu Jiang <xinyuj2@andrew.cmu.edu>
|
2026-06-20 01:25:01 -07:00 |
|
|
Xiaoyu Zhang
|
c1416bb3ee
|
[Diffusion] Keep FastHunyuan VAE resident on high-memory GPUs (#28773)
|
2026-06-20 15:54:06 +08:00 |
|
|
Lianmin ZhengandYinghai Lu
|
45d203fb08
|
Fix tokenizer state cleanup on dispatch failure (#28694)
Co-authored-by: Yinghai Lu <yinghai@meta.com>
|
2026-06-19 21:55:39 -07:00 |
|
|
Bi Xue
|
28e2096d1c
|
[sgl] wire SGLANG_OPT_SWA_RELEASE_LEAF_LOCK_AFTER_WINDOW on Unified Cache. (#28161)
|
2026-06-20 12:06:52 +08:00 |
|
|
Jimmy Shong
|
7516f0db9f
|
[cookbook] Laguna-M.1: add PD disaggregation section (#28737)
|
2026-06-19 19:49:49 -07:00 |
|
|
Michael
|
653f6735a0
|
[AMD] add nightly kv_canary JIT benchmark suite (#28611)
|
2026-06-19 19:46:34 -07:00 |
|
|
Jae B.andR0CKSTAR
|
2cbe1e6404
|
[Apple Silicon] [MLX] Fix MlxModelRunnerStub.initialize() signature desync with base (#28660)
Co-authored-by: R0CKSTAR <yeahdongcn@gmail.com>
|
2026-06-19 18:05:22 -07:00 |
|
|
Michael
|
871ed0dc0c
|
Revert "ci: add 4-GPU mi35x runner and rebalance off the saturated 8-GPU pool" (#28751)
|
2026-06-19 16:23:18 -07:00 |
|
|
Yanbin Jiang
|
6b945c16f4
|
[LoRA] Fix experimental fast-path multi-adapter correctness + flashinfer 0.6.12 compatibility (#28091)
|
2026-06-19 16:20:19 -07:00 |
|
|
Kangyan-ZhouandClaude Opus 4.8
|
364bf976be
|
[router] Tokenize prompt once at ingress; forward input_ids to the engine (all policies) (#28744)
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
2026-06-19 16:07:17 -07:00 |
|
|
Shu Wang
|
c3bae61e16
|
[Bug] fix(DummyModelLoader): run post_load_weights before process_weights_after_loading (#28665)
|
2026-06-19 22:57:45 +00:00 |
|
|
Kangyan-ZhouandClaude Opus 4.8
|
abf7011cdd
|
[router] Raise chat body cap to 5 MiB for long contexts (#28742)
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
2026-06-19 15:50:34 -07:00 |
|
|
Michael
|
420004827c
|
[AMD] register 3 tests to stage-b-test-1-gpu-large-amd (batch-6) (#28736)
|
2026-06-19 15:45:16 -07:00 |
|
|
Michaelandmichaelzhang-ai
|
13aab2fc06
|
ci: add 4-GPU mi35x runner and rebalance off the saturated 8-GPU pool (#28745)
Co-authored-by: michaelzhang-ai <michaelzhang-ai@users.noreply.github.com>
|
2026-06-19 15:41:19 -07:00 |
|
|
Mohammad Miadh Angkad
|
3a574846ff
|
[CI] Publish 4-GPU nightly profiler traces (#28738)
|
2026-06-19 15:27:24 -07:00 |
|
|
Elizaveta MartirosianandElizaveta Martirosian
|
b225f48fcc
|
[NPU][FIX CI] Set the warm-up mode to "request" for NPU diffusion tests (#28724)
Co-authored-by: Elizaveta Martirosian <elizaveta.martirosian@gmail.com>
|
2026-06-20 01:14:26 +03:00 |
|
|
|
3ed46f599f
|
[core] Don't force seq_lens_cpu publication under piecewise CUDA graph (#28633)
Co-authored-by: jonnykong <jonnykong@fb.com>
Co-authored-by: Liangsheng Yin <hnyls2002@gmail.com>
Co-authored-by: hnyls2002 <lsyincs@gmail.com>
|
2026-06-19 15:12:07 -07:00 |
|
|
Liangsheng Yin
|
d271de64fe
|
[misc] Move bench_one_batch_server into sglang/benchmark/ with a back-compat shim (#28625)
|
2026-06-19 14:19:29 -07:00 |
|
|
Liangsheng Yin
|
c9c2445146
|
[CI] Bump actions/github-script to v8 (Node 24 runtime) (#28741)
|
2026-06-19 14:18:37 -07:00 |
|
|
Cheng Wan
|
6fdcb9934c
|
fix(runner): size eager static buffers for prefill budget and MLP-sync autotune (#28677)
|
2026-06-19 13:23:26 -07:00 |
|
|
Cheng Wan
|
2aa7b58aa7
|
refactor(runner): reuse a prepared static buffer for every dummy run (#28740)
|
2026-06-19 13:16:07 -07:00 |
|
|
Cheng WanandClaude Opus 4.8
|
856b0dc74b
|
refactor(runner): move kernel warmup into the shared runner lifecycle (warmup()) (#28739)
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
2026-06-19 13:13:13 -07:00 |
|