 Binyao Jiang
|
6092721594
|
[Piecewise] Fix PCG issue for multimodal and embedding model that wraps language_model (#17290)
|
2026-01-20 14:06:06 -08:00 |
|
|
Binyao Jiang
|
38c233fd04
|
[Piecewise] Support PCG weak_ref_tensor cuda kernel on AMD (#17291)
|
2026-01-20 14:05:32 -08:00 |
|
|
Binyao Jiang
|
9a9f996f8d
|
[FP8] Fix weight_scale shape to match with x_scale shape for per-tensor quant under torch.compile (#16356)
|
2026-01-05 13:55:26 -08:00 |
|
|
Binyao Jiang
|
312df1d6c0
|
Fix TestGLM41VPPAccuracy test flakiness (#14848)
|
2025-12-10 16:59:58 -08:00 |
|
|
Binyao Jiang
|
a4992873d4
|
Treat unittest SkipTest exception as pass instead of as failure (#14847)
|
2025-12-10 15:28:21 -08:00 |
|
|
Binyao Jiang
|
ef3f8c97e1
|
Add ffmpeg into sglang docker - required by transformers multimodal V… (#14679)
|
2025-12-08 18:00:23 -08:00 |
|
|
Binyao Jiang
|
6abb8051e8
|
Bump up diffusers to latest official release version (#14670)
|
2025-12-08 13:41:01 -08:00 |
|
|
|
cf0478d602
|
[Glm46v] Bug fix for accuracy drop and unable to launch server (#14585)
Co-authored-by: yhyang201 <yhyang201@gmail.com>
Co-authored-by: zRzRzRzRzRzRzR <2448370773@qq.com>
Co-authored-by: Minglei Zhu <mingleizhu1122@gmail.com>
|
2025-12-07 23:45:02 -08:00 |
|
|
|
0f8e53947d
|
[Piecewise] Use same global graph memory pool as the main cuda graph … (#14044)
Co-authored-by: Stefan He <hebiaobuaa@gmail.com>
Co-authored-by: BBuf <1182563586@qq.com>
|
2025-11-30 23:04:10 -08:00 |
|
|
Binyao Jiang
|
de430b6745
|
[Performance] Replace preprocess_video logic from GLM multimodal processor with transformer impl for speed up (up to 27% faster) and addressing OOM (up to 50x improvements) (#13487)
|
2025-11-24 18:17:13 -08:00 |
|
|
Binyao Jiang
|
b29769f3b6
|
Move unnecessary input_addr capture under debug mode flag for speed-up (#13690)
|
2025-11-22 11:42:26 -08:00 |
|
|
Binyao Jiang
|
67fca6b297
|
[GDN] Remove unnecessary conv state clone (#13603)
|
2025-11-19 16:09:12 -08:00 |
|
|
Binyao Jiang
|
f88b2aa6af
|
[GDN] Remove unnecessary contiguous() (#13604)
|
2025-11-19 16:07:49 -08:00 |
|
|
Binyao Jiang
|
26ca07469b
|
[GLM4.6v] Relax the constraint of non-user role chat completion message schema for new GLM-v release (#13258)
|
2025-11-18 10:58:17 +08:00 |
|
|
Binyao Jiang
|
90c18a16cb
|
[GLM4.6v] Required changes for bumping up to transformer 5.x (#13229)
|
2025-11-18 10:58:00 +08:00 |
|
|
Binyao Jiang
|
7cdaedb8fb
|
Remove glm41v from CI to speed up CI (#13257)
|
2025-11-13 21:14:04 -08:00 |
|
|
Binyao Jiang
|
9b41f31a66
|
Use 32x32 black image for VLM server warmup and bring glm4.1v back to UT (#13222)
|
2025-11-13 14:21:43 -08:00 |
|
|
Binyao Jiang
|
9db3add319
|
Update GDN causal conv1d cuda kernel - prepare for new changes (#13188)
|
2025-11-13 14:09:47 -08:00 |
|
|
Binyao Jiang
|
fbbe16faab
|
[GDN] Fuse b.sigmoid(), fused_gdn_gating and unsqueeze into one kernel: up to 0.85% e2e speedup (#12508)
|
2025-11-05 22:52:04 -08:00 |
|
|
Binyao JiangandStefan He
|
3451fc3280
|
[Feature] Qwen3-Next & FLA: Support MTP topk>1; Up to 6% faster (#11133)
Co-authored-by: Stefan He <hebiaobuaa@gmail.com>
|
2025-11-01 19:47:56 -07:00 |
|
|
Binyao Jiang
|
0aa65f94f1
|
[Fix] Improve longbench prompt and other logics (#11474)
|
2025-10-12 15:04:28 -07:00 |
|
|
Binyao Jiang
|
451d15c44b
|
[DPSKv3.2] Rewrite nsa tilelang act_quant kernel to triton (#11450)
|
2025-10-10 23:13:46 -07:00 |
|
|
Binyao Jiang
|
bd7eb0205a
|
[Performance] Qwen3-Next: optimize causal_conv1d_fn triton kernel - up to 9% faster (#10680)
|
2025-09-20 00:12:43 -07:00 |
|
|
Binyao Jiang
|
4e6c4923a0
|
[Performance] Qwen3-Next: speed up update_mamba_state_after_mtp_verify by 10x; e2e up to 3.54% faster (#10586)
|
2025-09-18 17:13:59 -07:00 |
|
|
Binyao Jiang
|
b91cb67e7a
|
[Performance] Qwen3-Next: replace arange to cached query_start_loc_li… (#10553)
|
2025-09-18 17:02:42 -07:00 |
|
|
Binyao Jiang
|
9752861002
|
[Fix] Support qwen3-next MTP+DP (#10392)
|
2025-09-13 17:45:04 +08:00 |
|
|
Binyao Jiang
|
31e9d3a5aa
|
[Fix] Init mamba related memory pools with torch.zeros (#10400)
|
2025-09-13 14:16:48 +08:00 |
|
|
Binyao Jiang
|
24a8cee66d
|
Fix GLM45v launch server cuda torch compile bug (#9554)
|
2025-08-25 13:46:28 -07:00 |
|
|
Binyao Jiang
|
3affa9dcc3
|
Fix GLM45 tool call multi-turn bug (#9500)
|
2025-08-25 13:46:13 -07:00 |
|
|
Binyao Jiang
|
c2fbf60f39
|
[GLM4.1V and GLM4.5V] Add vision transformer num_dummy_head support: max tp=4 -> max tp=8 (#9059)
|
2025-08-18 14:40:13 -07:00 |
|
|
Binyao Jiang
|
66d6be0874
|
Bug fix: use correct mm_items in embed_mm_inputs (#8893)
|
2025-08-16 19:55:56 -07:00 |
|
|
|
f29aba8c6e
|
Support glm4.1v and glm4.5v (#8798)
Signed-off-by: Xinyuan Tong <justinning0323@outlook.com>
Signed-off-by: Xinyuan Tong <xinyuantong.cs@gmail.com>
Co-authored-by: Xinyuan Tong <justinning0323@outlook.com>
Co-authored-by: Xinyuan Tong <115166877+JustinTong0323@users.noreply.github.com>
Co-authored-by: Xinyuan Tong <xinyuantong.cs@gmail.com>
Co-authored-by: zRzRzRzRzRzRzR <2448370773@qq.com>
Co-authored-by: Minglei Zhu <mingleizhu1122@gmail.com>
Co-authored-by: Chang Su <csu272@usc.edu>
|
2025-08-09 00:59:13 -07:00 |
|
|
Binyao Jiang
|
7b81f956eb
|
Fix qwen2 audio not working bug (#8600)
|
2025-08-09 00:42:29 -07:00 |
|
|
Binyao JiangandCheng Wan
|
c8d3a402c1
|
Bug: apply final_hidden_states*=self.routed_scaling_factor at MoE lay… (#8511)
Co-authored-by: Cheng Wan <54331508+ch-wan@users.noreply.github.com>
|
2025-08-01 00:07:41 -07:00 |
|
|
Binyao Jiang
|
59aab76f0a
|
Bug: Fix google gemma3n-mm audio input not working bug (#8365)
|
2025-07-30 21:23:09 -07:00 |
|
|
Binyao Jiang
|
581e7dcb92
|
GLM-4.5 Model Support Follow-up (#8445)
|
2025-07-27 23:35:20 -07:00 |
|
|
Binyao JiangandChang Su
|
e983d66680
|
Fix: Improve test_openai_function_calling unit test and fix reasoning_parser.py think_start_token logic (#8316)
Co-authored-by: Chang Su <chang.s.su@oracle.com>
|
2025-07-27 13:12:59 -07:00 |
|
|
Binyao Jiang
|
b7e951a6db
|
Feat: Support audio in Phi4-mm model (#8048)
|
2025-07-18 21:03:53 -07:00 |
|
|
Binyao Jiang
|
2d54d4bb64
|
Feat: Support Phi-3.5-MoE in SGLang (#7907)
|
2025-07-09 23:51:33 -07:00 |
|
|
Binyao Jiang
|
b783c1cb82
|
Fix hicache benchmark script bug - some sampled input_request is [] (#7300)
|
2025-06-17 23:47:11 -07:00 |
|