Commit Graph

1563 Commits

Author SHA1 Message Date
Mick
ff2ce0b86f refactor: move image processors to separate files (#4229) 2025-03-11 12:35:35 -07:00
Ximingwang-09
0f2a2e3c19 Add H20 tuning configs support DeepSeek V3/R1 INT8(block-wise) (#4220)
Co-authored-by: ximing.wxm <ximing.wxm@antgroup.com>
2025-03-11 12:32:33 -07:00
yigex
690e1f2371 [AMD] Fix rocm sgl-kernel missing modules error (#4311)
Co-authored-by: yiakwy-xpu-ml-framework-team <leiwang2@amd.com>
2025-03-11 10:35:28 -07:00
yych0745
6a02b32d07 Add A100 tuning configs for DeepSeek R1/V3 channel-wise INT8 (#4287)
Co-authored-by: HandH1998 <1335248067@qq.com>
2025-03-11 00:49:06 -07:00
lukec
dce303e279 linear support deepgemm (#4199)
Co-authored-by: yinfan98 <1106310035@qq.com>
2025-03-11 00:38:37 -07:00
Yineng Zhang
4d27eb9ad1 update sgl-kernel 0.0.4.post2 (#4291) 2025-03-11 00:34:33 -07:00
lambert0312
d3ecd63204 Add A800 tuning configs support DeepSeek V3/R1 BF16 and INT8(block-wise) (#4136) 2025-03-11 00:32:25 -07:00
Yineng Zhang
e187a3d595 upgrade xgrammar 0.1.15 (#4275) 2025-03-10 14:53:24 -07:00
HandH1998
2ac189edc8 Amd test fp8 (#4261) 2025-03-10 10:12:09 -07:00
Lianmin Zheng
5a6400eec5 Test no vllm custom allreduce (#4256) 2025-03-10 10:08:25 -07:00
Lianmin Zheng
00d25a7f5e Fix quantization and nightly tests (#4258) 2025-03-10 03:06:21 -07:00
shimin
ac69885056 fix the input_ids is None error (#4144) 2025-03-10 01:38:37 -07:00
Lianmin Zheng
aa957102a9 Simplify tests & Fix trtllm custom allreduce registration (#4252) 2025-03-10 01:24:22 -07:00
DavidChan
4455b26e76 [Bug fixed] fixed the crash when enable the dp-attention on the single card (#3958) 2025-03-10 00:50:34 -07:00
Lianmin Zheng
e8a69e4d0c Clean up fp8 support (#4230) 2025-03-09 21:46:35 -07:00
Lianmin Zheng
fbd560028a Auto balance CI tests (#4238) 2025-03-09 21:05:55 -07:00
Lianmin Zheng
730d084f2a Minor style fix for sgl-kernel (#4243) 2025-03-09 20:15:13 -07:00
Lianmin Zheng
4a05bdfa86 Revert "Check eagle server args" (#4242) 2025-03-09 18:53:33 -07:00
Ying Sheng
34c8898755 Check eagle server args (#4217) 2025-03-09 01:10:43 -08:00
HandH1998
0dd6cda288 Apply sgl w8a8 fp8 kernel (#3148) 2025-03-09 00:03:32 -08:00
Baizhou Zhang
9fb48f951f Support nextn for flashinfer mla attention backend (#4218) 2025-03-09 00:01:54 -08:00
Yineng Zhang
89ccb533ad use sgl-kernel 0.0.4 (#4224) 2025-03-08 23:43:09 -08:00
Lianmin Zheng
1361ab9e03 Lazily import lora backends (#4225) 2025-03-08 23:39:26 -08:00
Lianmin Zheng
8abf74e3c9 Rename files in sgl kernel to avoid nested folder structure (#4213)
Co-authored-by: zhyncs <me@zhyncs.com>
2025-03-08 22:54:51 -08:00
Mingshan
0fe7c13be1 Fix bench_serving flush cache not recognizing OPENAI_API_KEY (#4181)
Signed-off-by: Mingshan <git@brighill.com>
2025-03-08 01:03:38 -08:00
Lianmin Zheng
08c4d764a5 lazy import attn backends (#4200) 2025-03-08 00:41:35 -08:00
Lianmin Zheng
d4017a6b63 [EAGLE] many fixes for eagle (#4195)
Co-authored-by: SangBin Cho <rkooo567@gmail.com>
Co-authored-by: Sehoon Kim <sehoon@x.ai>
2025-03-07 22:12:13 -08:00
Lianmin Zheng
d052f4c8a9 New clang format for sgl kernel (#4194) 2025-03-07 20:21:08 -08:00
Ke Bao
20c8119915 Fix eagle hang issue for max_new_tokens=1 (#4185) 2025-03-07 12:11:18 -08:00
Yineng Zhang
eb61f5c9af Revert "ROCm: Flex Attention Enablement with custom backends (#4178)" (#4186) 2025-03-07 10:27:52 -08:00
HAI
0beea4503f ROCm: Flex Attention Enablement with custom backends (#4178)
Co-authored-by: linsun12 <linsun12@amd.com>
2025-03-07 04:38:53 -08:00
Zhiqiang Xie
9376ac361d Memory pool fix for upstream change about eagle (#4170) 2025-03-07 00:58:20 -08:00
HandH1998
c7f254468f [Feature] DeepSeek V3/R1 INT8 Quantization (channel-wise) (#3888)
Co-authored-by: yych0745 <1398089567@qq.com>
Co-authored-by: sleepcoo <sleepcoo@gmail.com>
Co-authored-by: b0urnee <2769086541@qq.com>
2025-03-06 20:54:52 -08:00
Pan Lyu
361971b859 Add Support for Qwen2-VL Multi-modal Embedding Models (#3694) 2025-03-06 16:46:20 -08:00
HAI
13bc39c5d6 ROCm: enable trillion-parameter MoE models with INT4-FP8 single node (#4152) 2025-03-06 15:33:02 -08:00
Lianmin Zheng
9c58e68b4c Release v0.4.3.post4 (#4140) 2025-03-06 12:50:28 -08:00
Oliver Stanley
d03b3467b8 Fix constrained generation errors by adding datasets dependency (#4142) 2025-03-06 12:07:51 -08:00
yinfan98
ab7fba0ece Fix nightly ci Gsm8k & Fix flashinfer backend kvcache quant (#4147) 2025-03-06 11:50:07 -08:00
Lianmin Zheng
bc1534ff32 Fix a draft model accuracy bug in eagle; support step=1; return logprob in eagle (#4134)
Co-authored-by: Sehoon Kim <kssteven418@gmail.com>
Co-authored-by: SangBin Cho <rkooo567@gmail.com>
Co-authored-by: Sehoon Kim <sehoon@x.ai>
2025-03-06 06:13:59 -08:00
Lzhang-hub
3a3918121f fix bench serving bug (#4135) 2025-03-06 05:34:02 -08:00
Lianmin Zheng
98c73d71cb [Minor] make the __init__ function of model_runner.py shorter (#4132) 2025-03-06 01:51:12 -08:00
Lianmin Zheng
fcc2e37f69 Split the __init__ of scheduler as smaller functions. Improve the eagle tests (#4128) 2025-03-06 00:13:20 -08:00
Zhiqiang Xie
aee30630d8 Add a pointer to the real KV cache pool (#4113) 2025-03-05 21:39:07 -08:00
Lianmin Zheng
286e6540a6 Remove prefill-only-one-req (#4117) 2025-03-05 20:58:48 -08:00
Wenxuan Tan
718c391fd7 [Hoxfix] Fix incomplete token_to_kv_pool refactor (#4121) 2025-03-05 19:32:42 -08:00
Yineng Zhang
fc671f66c1 chore: bump v0.4.3.post3 (#4114) 2025-03-05 17:26:10 -08:00
Yueyang Pan
25482edb5c Online serving benchmarks of real datasets for hierarchical KV caching (#3211)
Co-authored-by: Zhiqiang Xie <xiezhq@stanford.edu>
2025-03-05 16:16:43 -08:00
luzengxiangcn
62b362b1f1 Debug radixcache: refactor recursive helper methods (#3029)
Co-authored-by: Zhiqiang Xie <xiezhq@stanford.edu>
2025-03-05 16:11:42 -08:00
Jhin
70b3c6eeb1 Add update_weights_from_disk endpoint to Engine (#4102)
Co-authored-by: zhaochenyang20 <zhaochen20@outlook.com>
2025-03-05 12:25:18 -08:00
Ke Bao
ef9d3b3c2c Fix triton kernel illegal memory issue for eagle (#4100) 2025-03-05 11:23:53 -08:00