EngineX/xc-llm-kunlun
3
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
122 Commits 2 Branches 0 Tags
d18df18499c5acb277a8132bad44b53d7e41de4d
Commit Graph

122 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
WeiJie_Hong
d18df18499 [CI/Build] update .pre-commit-config.yaml && add _pylint.yml && update installation.md (#155) 
Signed-off-by: WeiJie_Hong <1462519292@qq.com>
 2026-01-28 17:58:46 +08:00
Li Wei
71bd70ad6c [Feature] support compressed-tensors w4a16 quantization (#154) 
- native int4 kimi model inference is supported

Signed-off-by: Li Wei <liwei.109@outlook.com>
 2026-01-27 19:56:22 +08:00
Shiwen Tang
0711c1abfa [Feature] Support AWQ MoE W4A16 Quantization (#142) 
Signed-off-by: tangshiwen <tangshiwen@baidu.com>
Co-authored-by: Li Wei <liwei.109@outlook.com>
 2026-01-26 18:56:05 +08:00
WeiJie_Hong
2a998286c0 [Doc] update base image url（1.Replace conda with uv; 2.Integrate xpytorch and ops into the image.) (#146) 
Signed-off-by: WeiJie_Hong <1462519292@qq.com>
 2026-01-23 18:55:56 +08:00
1916hcc
c0f06d04b1 [Doc] docs: remove internal pip index from requirements (#147) 
Signed-off-by: Chenchao Hu <huchenchao@example.com>
Co-authored-by: Chenchao Hu <huchenchao@example.com>
 2026-01-23 18:55:34 +08:00
baoqian426
1eaa1336ac [Bugfix]remove mla patch, server args no need --compilation-config for ds v3.1 (#145) 
Signed-off-by: baoqian426 <1354987947@qq.com>
 2026-01-23 15:59:43 +08:00
fromck
0ce5f1a3f7 Add kernels to optimize RoPE and the decoding stage (#143) 
Co-authored-by: chengxiaokang <chengxiaokang@baidu.com>
 2026-01-23 10:29:52 +08:00
Lidang Jiang
9e13f23661 [Doc] Optimize the document (#136) 2026-01-22 14:12:44 +08:00
Joeegin
58f570ddea [Docs] Add XPU tutorials for Qwen / InternVL (#140) 
Signed-off-by: Joeegin <3318329726@qq.com>
 2026-01-22 13:50:49 +08:00
fromck
74d4f804e8 add 2 kernels and optimize the calculation of topk_indices (#134) 
Co-authored-by: chengxiaokang <chengxiaokang@baidu.com>
 2026-01-22 10:29:28 +08:00
yuqilinaa
c9f00c132c [Kernel] Enable fast random sample on Kunlun3 Platform with generators (#73) 
Co-authored-by: Xinyu Dong <dongxinyu03@baidu.com>
 2026-01-20 21:49:33 +08:00
WANG HAO
c404af3a41 [Feature] totaly support multi-lora support,latest xspeedgate needed (#133) 
Co-authored-by: wanghao <wanghao@example.com>
 2026-01-20 21:27:02 +08:00
youzeyu
92b40628cd delete glmGlmForCausalLM register (#132) 
Co-authored-by: hanhaowen <hanhaowen@baidu.com>
 2026-01-20 19:22:33 +08:00
haoli5009-debug
561a235a3f [CI/Build] Modify biweekly report readme files (#131) 
Co-authored-by: v_lihao66 <v_lihao66@baidu.com>
 2026-01-20 16:58:36 +08:00
Li Wei
2a2d773ad0 [fix]bias bug in kunlun_scale_mm (#126) 2026-01-20 13:24:52 +08:00
Li Wei
f2019b145f Revert "support glm47 in 0.11.0 version (#116)" (#123) 
This reverts commit 9006e37979.
 2026-01-20 10:46:11 +08:00
roger-lcc
9006e37979 support glm47 in 0.11.0 version (#116) 
* support glm47 in 0.11.0 version

* support glm47 in 0.11.0 version

---------

Co-authored-by: luochencheng <luochencheng@baidu.com>
 2026-01-19 20:26:26 +08:00
Li Wei
8f56cbf3ed [refactor]update Kunlun classes with monkey patch (#122) 
Signed-off-by: Li Wei <liwei.109@outlook.com>
 2026-01-19 20:24:19 +08:00
baoqian426
2512259944 longcontext chunk make attention crash, fix it (#117) 
Co-authored-by: root <root@rdtest-node1150.bcc-zwlt.baidu.com>
 2026-01-17 18:38:23 +08:00
fromck
71a5a04e0a [Misc]Specify that DS32 only supports --kv-cache-dtype bfloat16 (#119) 
* [Kernel] add kernels to torch.ops

* [Misc]Specify that DS only supports --kv-cache-dtype bfloat16

---------

Co-authored-by: chengxiaokang <chengxiaokang@baidu.com>
 2026-01-17 16:52:02 +08:00
Shiwen Tang
8988ad08b2 [Feature] Support Mixed-Precision Quantization for MoE (#112) 2026-01-14 18:42:18 +08:00
baoqian426
6706651646 Merge pull request #91 from zhihui96/dsv31 
[Feature]enable int8 bmm
 2026-01-14 15:12:30 +08:00
wzh
115eb32068 enable int8 bmm 2026-01-14 14:30:59 +08:00
zhihui96
f0bf384e2e Merge branch 'baidu:main' into dsv31 2026-01-14 14:21:57 +08:00
Lidang Jiang
7ed71432ca [Bug] Fix InternVL KeyError: ((1, 1, 3), '<i8') (#108) 2026-01-13 22:36:03 +08:00
roger-lcc
37cc307322 register apply_repetition_penalties_ in custom_op (#110) 
* fix qwen2_vl for 0.11.0

* register apply_repetition_penalties_ in custom_op

---------

Co-authored-by: luochencheng <luochencheng@baidu.com>
 2026-01-13 20:22:14 +08:00
baoqian426
fb424acca7 Merge pull request #106 from baoqian426/enable-full-cudagraph-deepseek 
enable full cudagraph for deepseek
 2026-01-13 09:57:56 +08:00
Jin Hanyu
bd90350968 [Bug] Fix no apply_top_k_top_p issue. (#101) 2026-01-12 16:38:03 +08:00
tanjunchen
18fc1c006e update maintainer for vllm-kunlun (#100) 
Signed-off-by: tanjunchen <tanjunchen20@gmail.com>
 2026-01-12 16:37:22 +08:00
hanhaowen
ff8ebfa208 enable full cudagraph for deepseek 2026-01-12 15:18:12 +08:00
Xinyu Dong
87a57e43ca [Docs] Upate URL (#98) 2026-01-10 06:02:10 +08:00
Xinyu Dong
7be26ca617 [Bugs] Fix Docs Build Problem (#97) 
* [Bugs] Docs fixed

* Update contributing.md

* Update index.md

* fix lua to text

* fix title size
 2026-01-10 05:55:40 +08:00
baoqian426
8c9cabd760 Merge pull request #96 from xyDong0223/main 
[Docs] Fix v0.11.0 Docs config
 2026-01-09 17:20:17 +08:00
Xinyu Dong
462c44e2ac [Docs] Fix v0.11.0 Docs config 2026-01-09 17:07:18 +08:00
roger-lcc
0455b49519 [Bugs] fix qwen2_vl for 0.11.0 (#94) 
Co-authored-by: luochencheng <luochencheng@baidu.com>
 2026-01-09 15:05:40 +08:00
wzh
df436a47f6 test 2026-01-08 16:02:15 +08:00
baoqian426
2c9b176e6e [Feature] use for dp (#90) 2026-01-08 11:05:48 +08:00
Li Wei
c403d921ff [doc] update quantization guide doc (#88) 2026-01-07 15:39:51 +08:00
baoqian426
eb40e8a07a [Bugfix] fix can not import compressed_tensors (#87) 
Co-authored-by: root <root@rdtest-node1150.bcc-zwlt.baidu.com>
 2026-01-07 11:32:10 +08:00
baoqian426
62a97db6ed Merge pull request #85 from liwei109/liwei-dev 
[fix] update compressed-tensors scheme
 2026-01-07 09:27:23 +08:00
Li Wei
1c1b84d78c [fix]update compressed-tensors scheme 
Deepseek v3.2 is supported now

Signed-off-by: Li Wei <liwei.109@outlook.com>
 2026-01-06 22:30:27 +08:00
baoqian426
9c2b908908 Merge pull request #84 from xyDong0223/main 
[Feature] DeepSeek Support MTP
 2026-01-06 21:56:31 +08:00
baoqian426
c5e4d23e3e Merge pull request #82 from liwei109/quant 
[fix] resolve cutlass_scaled_mm inference error
 2026-01-06 21:42:55 +08:00
dongxinyu03
26b311ccf5 [Feature] DeepSeek Support MTP 2026-01-06 21:37:21 +08:00
tangshiwen
f811ae968a [fix] resolve cutlass_scaled_mm inference error 2026-01-06 20:52:12 +08:00
baoqian426
c54b2d2a2d Merge pull request #80 from liwei109/aicapx-quant 
[fix]matmul not support cuda graph
 2026-01-06 17:49:09 +08:00
Li Wei
9533f68e99 [fix]matmul not support cuda graph 2026-01-06 17:32:45 +08:00
Li Wei
515a4eeda9 [dev] support compressed-tensors w8a8 quantization (#75) 
* [dev] support compressed-tensors w8a8 quantization

Co-authored-by: Li Wei <liwei.109@outlook.com>

* [refact]update KunlunScaleMMKernel impl

* [rebase]resolve conflicts and remove redundant code

---------

Co-authored-by: tangshiwen <tangshiwen@baidu.com>
 2026-01-06 13:51:53 +08:00
baoqian426
ee0f50e68f [Feature] support deepseek v3/r1/v3.2 (#78) 
* [Feature] support deepseek v3/r1/v3.2

* fix gpt_oss

* update readme

* update readme

---------

Co-authored-by: hanhaowen <hanhaowen@baidu.com>
 2026-01-05 22:55:35 +08:00
Xinyu Dong
07bc24a555 [Bugs] Fix moe when without bias (#76) 2026-01-05 10:51:23 +08:00