Commit Graph

3231 Commits

Author SHA1 Message Date
kevin
753772ace8 default enable chunked prefill (#3731)
* add error traceback info

* update error msg

* update code

* default enable chunked prefill

* update code

* update code

* add envs

* update code

---------

Co-authored-by: Jiang-Jia-Jun <163579578+Jiang-Jia-Jun@users.noreply.github.com>
2025-08-31 13:15:13 +08:00
李泳桦
98e03fb4ea [feat] add metrics for yiyan adapter (#3219) (#3614)
Some checks failed
CE Compile Job / ce_job_pre_check (push) Has been cancelled
CE Compile Job / print_ce_job_pre_check_outputs (push) Has been cancelled
CE Compile Job / FD-Clone-Linux (push) Has been cancelled
CE Compile Job / Show Code Archive Output (push) Has been cancelled
CE Compile Job / BUILD_SM8090 (push) Has been cancelled
CE Compile Job / BUILD_SM8689 (push) Has been cancelled
CE Compile Job / CE_UPLOAD (push) Has been cancelled
Deploy GitHub Pages / deploy (push) Has been cancelled
Publish Job / publish_pre_check (push) Has been cancelled
Publish Job / print_publish_pre_check_outputs (push) Has been cancelled
Publish Job / FD-Clone-Linux (push) Has been cancelled
Publish Job / Show Code Archive Output (push) Has been cancelled
Publish Job / BUILD_SM8090 (push) Has been cancelled
Publish Job / BUILD_SM8689 (push) Has been cancelled
Publish Job / PADDLE_PYPI_UPLOAD_8090 (push) Has been cancelled
Publish Job / PADDLE_PYPI_UPLOAD_8689 (push) Has been cancelled
Publish Job / Run FastDeploy Unit Tests and Coverage (push) Has been cancelled
Publish Job / Run FastDeploy LogProb Tests (push) Has been cancelled
Publish Job / Extracted partial CE model tasks to run in CI. (push) Has been cancelled
Publish Job / Run Base Tests (push) Has been cancelled
Publish Job / Run Accuracy Tests (push) Has been cancelled
* [feat] add metrics for yiyan adapter

* [fix] fix metrics num_requests_waiting and num_requests_running

* [fix] fix metrics gpu_cache_usage_perc

* [refactor] change where requests_number increases

* [chore] rename xxx_block_num as xxx_gpu_block_num, and update their values accordingly

* [chore] delete useless code
2025-08-30 23:20:58 +08:00
Sunny-bot1
fe5d09f9ee [FIX]Fix Machete compile via ENABLE_MACHETE (#3727)
* add ENABLE_MACHETE

* fix

* revert

* update

* pre_commit

* fix

* fix

---------

Co-authored-by: Ayakouji <yuhongh@qq.com>
Co-authored-by: Jiang-Jia-Jun <163579578+Jiang-Jia-Jun@users.noreply.github.com>
Co-authored-by: aquagull <hongyuh@qq.com>
2025-08-30 17:50:17 +08:00
SunLei
b9af95cf1c [Feature] Add AsyncTokenizerClient&ChatResponseProcessor with remote encode&decode support. (#3674)
* [Feature] add AsyncTokenizerClient

* add decode_image

* Add response_processors with remote decode support.

* [Feature] add tokenizer_base_url startup argument

* Revert comment removal and restore original content.

* [Feature] Non-streaming requests now support remote image decoding.

* Fix parameter type issue in decode_image call.

* Keep completion_token_ids when return_token_ids = False.

* add copyright
2025-08-30 17:06:26 +08:00
luukunn
9a7c231f2c [Feature]support chat_template.jinja (#3721)
* add support chat_template.jinja

* add support chat_template.jinja
2025-08-30 17:05:34 +08:00
lizexu123
b21e085f3e [Code Simplification] delete print (#3729) 2025-08-30 16:19:07 +08:00
chen
7568b20098 check (#3720) 2025-08-30 16:04:20 +08:00
lizexu123
455205f991 [Features] support hugging face qwen3 moe (#3649)
Some checks failed
CE Compile Job / ce_job_pre_check (push) Has been cancelled
CE Compile Job / print_ce_job_pre_check_outputs (push) Has been cancelled
CE Compile Job / FD-Clone-Linux (push) Has been cancelled
CE Compile Job / Show Code Archive Output (push) Has been cancelled
CE Compile Job / BUILD_SM8090 (push) Has been cancelled
CE Compile Job / BUILD_SM8689 (push) Has been cancelled
CE Compile Job / CE_UPLOAD (push) Has been cancelled
Deploy GitHub Pages / deploy (push) Has been cancelled
* split ut

* qwen3-30B-A3B

* fix

* add test

* add test_torch_model.py

* fix test_torch_model.py

* delete print

* fix moe

* delete init.py

* fix

* fix

---------

Co-authored-by: bukejiyu <395822456@qq.com>
Co-authored-by: bukejiyu <52310069+bukejiyu@users.noreply.github.com>
2025-08-30 15:26:05 +08:00
Zero Rains
f206474cc7 fix the bug when num_key_value_heads < tensor_parallel_size (#3717) 2025-08-30 12:40:00 +08:00
chenjian
c4b1f6b0a5 [Optimize] Increase zmq buffer size to prevent apiserver too slowly to consume (#3723) 2025-08-30 10:45:26 +08:00
YUNSHEN XIE
a18afcfdd9 Optimize coverage jobs (#3683)
Some checks failed
CE Compile Job / ce_job_pre_check (push) Has been cancelled
CE Compile Job / print_ce_job_pre_check_outputs (push) Has been cancelled
CE Compile Job / FD-Clone-Linux (push) Has been cancelled
CE Compile Job / Show Code Archive Output (push) Has been cancelled
CE Compile Job / BUILD_SM8090 (push) Has been cancelled
CE Compile Job / BUILD_SM8689 (push) Has been cancelled
CE Compile Job / CE_UPLOAD (push) Has been cancelled
Deploy GitHub Pages / deploy (push) Has been cancelled
2025-08-30 00:12:40 +08:00
chen
cd252ec673 [Feature]support load eb 0.3B and 21B torch model (#3660)
Some checks failed
CE Compile Job / ce_job_pre_check (push) Has been cancelled
CE Compile Job / print_ce_job_pre_check_outputs (push) Has been cancelled
CE Compile Job / FD-Clone-Linux (push) Has been cancelled
CE Compile Job / Show Code Archive Output (push) Has been cancelled
CE Compile Job / BUILD_SM8090 (push) Has been cancelled
CE Compile Job / BUILD_SM8689 (push) Has been cancelled
CE Compile Job / CE_UPLOAD (push) Has been cancelled
Deploy GitHub Pages / deploy (push) Has been cancelled
Publish Job / publish_pre_check (push) Has been cancelled
Publish Job / print_publish_pre_check_outputs (push) Has been cancelled
Publish Job / FD-Clone-Linux (push) Has been cancelled
Publish Job / Show Code Archive Output (push) Has been cancelled
Publish Job / BUILD_SM8090 (push) Has been cancelled
Publish Job / BUILD_SM8689 (push) Has been cancelled
Publish Job / PADDLE_PYPI_UPLOAD_8090 (push) Has been cancelled
Publish Job / PADDLE_PYPI_UPLOAD_8689 (push) Has been cancelled
Publish Job / Run FastDeploy Unit Tests and Coverage (push) Has been cancelled
Publish Job / Run FastDeploy LogProb Tests (push) Has been cancelled
Publish Job / Extracted partial CE model tasks to run in CI. (push) Has been cancelled
Publish Job / Run Base Tests (push) Has been cancelled
Publish Job / Run Accuracy Tests (push) Has been cancelled
2025-08-29 20:00:48 +08:00
yangjianfengo1
3754a9906d [Feature] block sparse attention (#3668)
* 支持稀疏attn

* fix bug

* code style

* fix moba attn get kv shape

* 修复a100编译

* codestyle

* code style

* code style

* code style

* fix conflict

* 增加单侧

* code style

* 增加eblite 加载时间

* fix bug

* for ci

* for ci

* for ci

* for ci

* 支持mlp block size 128

* 增加小算子单测

* fix 单测 mlp

* 将环境变量加入到config里面

* fix rollout config

* 修复显存

* add test server

* add test server

* fix mlp  最后一层使用full attn
2025-08-29 19:46:30 +08:00
zhouchong
ccd52b5596 [Model]support qwen2_5_vl (#3557)
* adapt qwen_2_5_vl model

* adapt qwen_2_5_vl VIT model

* adapt qwen2_5_vl images_embeds

* adapt qwen2_5_vl 3D rope

* adapt qwen2_5_vl 3D rope v2

* adapt qwen2_5_vl processor

* adapt qwen2_5_vl bypass resampler_model

* adapt qwen2_5_vl 绕过部分ernie逻辑

* adapt qwen2_5_vl 绕过部分ernie逻辑 v2

* adapt qwen2_5_vl 权重加载与命名修改

* adapt qwen2_5_vl 非必须think_end_id

* adapt qwen2_5_vl 区分多种模型的extract_vision_features

* fix:adapt qwen2_5_vl model

* adapt qwen2_5_vl norm

* adapt qwen2_5_vl  processor 更新

* adapt qwen2_5_vl image and video success

* adapt qwen2_5_vl 部分整理代码

* adapt qwen2_5_vl 支持多卡

* adapt qwen2_5_vl on latest develop

* adapt qwen2_5_vl RL

* adapt qwen2_5_vl 整理代码

* support noex rope3d

* adapt qwen2_5_vl add init.py

* adapt qwen2_5_vl add init.py v2

* adapt qwen2_5_vl remove space

* adapt qwen2_5_vl remove space v2

* adapt qwen2_5_vl pre-commit

* adapt qwen2_5_vl update

* adapt qwen2_5_vl pre-commit v2

* adapt qwen2_5_vl modify comments

* adapt qwen2_5_vl fix indentation

* adapt qwen2_5_vl fix indentation v2

---------

Co-authored-by: wangyafeng <wangyafeng@baidu.com>
Co-authored-by: xiaoxiaohehe001 <49090790+xiaoxiaohehe001@users.noreply.github.com>
Co-authored-by: CSWYF3634076 <58356743+CSWYF3634076@users.noreply.github.com>
2025-08-29 18:28:39 +08:00
YuBaoku
65425bf858 [CI] update paddle version to nightly (#3698) 2025-08-29 18:16:13 +08:00
Yuan Xiaolan
c71ee0831c add w4afp8 offline script (#3636) 2025-08-29 17:56:05 +08:00
zyfncg
f677c032c0 [CudaGraph] [SOT] Support spliting static graph into piecewise graph with cuda_graph (#3478)
* support spliting static graph into piecewise graph with cuda_graph

* Update fastdeploy/model_executor/graph_optimization/cudagraph_piecewise_backend.py

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

* fix merge conflict

* fix bug

---------

Co-authored-by: YuBaoku <49938469+EmmonsCurse@users.noreply.github.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
2025-08-29 16:28:01 +08:00
lzy
48d760539b fix deepcopy(tp_group) in spec (#3648) 2025-08-29 16:08:21 +08:00
Ryan
45f81b34f0 add dtype int32 (#3692)
Some checks failed
CE Compile Job / ce_job_pre_check (push) Has been cancelled
CE Compile Job / print_ce_job_pre_check_outputs (push) Has been cancelled
CE Compile Job / FD-Clone-Linux (push) Has been cancelled
CE Compile Job / Show Code Archive Output (push) Has been cancelled
CE Compile Job / BUILD_SM8090 (push) Has been cancelled
CE Compile Job / BUILD_SM8689 (push) Has been cancelled
CE Compile Job / CE_UPLOAD (push) Has been cancelled
Deploy GitHub Pages / deploy (push) Has been cancelled
2025-08-29 14:56:35 +08:00
xiaoxiaohehe001
1bf4fc7f36 support w4afp8 eplb (#3680) 2025-08-29 14:43:06 +08:00
Yuanle Liu
68f87240da fix key error in mm (#3702) 2025-08-29 14:35:12 +08:00
李泳桦
88297240e7 [feat] completion api supports passing input token ids in either prompt or prompt_token_ids (#3311)
* [feat] completion api supports passing input token ids in either `prompt` or `prompt_token_ids`

* [fix] update comment

* [fix] fix type error

* [test] add a unittest file for serving api test

* [test] try to fix ci error

* [chore] rename test function names

* [test] try to fix ci error

* [test] try to fix ci error

* [test] add tests for qwen
2025-08-29 14:19:42 +08:00
周周周
17b414c2df MoE Default use triton's blockwise fp8 in TP Case (#3678) 2025-08-29 11:07:30 +08:00
co63oc
b6edd15d55 fix scaled_gemm_f8_i4_f16_weight_quantize input (#3685) 2025-08-29 11:04:04 +08:00
Yuanle Liu
2fb2c0f46a fix MultimodalRegistry (#3699) 2025-08-29 11:01:30 +08:00
Echo-Nie
43d5bd62b4 【Hackathon 9th No.70】supplementary unit test for CPUPlatform and CUDAPlatform (#3580)
* 功能模块 CUDAPlatform、CPUPlatform 单测补充

* update the "is_cuda" to "is_cuda_and_available"

* fix pre-commit

---------

Co-authored-by: Tao Luo <luotao02@baidu.com>
2025-08-29 10:34:05 +08:00
lifulll
72094d4d82 enable dcu ci (#3402) 2025-08-29 10:23:08 +08:00
kevin
73d60fe64d update ci envs for structred output (#3687)
* add error traceback info

* update error msg

* update code

* update ci envs for structred output

---------

Co-authored-by: Jiang-Jia-Jun <163579578+Jiang-Jia-Jun@users.noreply.github.com>
2025-08-29 10:21:36 +08:00
bukejiyu
0b51b9c35b fix qwen3 235B tp 8 (#3697)
Some checks failed
CE Compile Job / ce_job_pre_check (push) Has been cancelled
CE Compile Job / print_ce_job_pre_check_outputs (push) Has been cancelled
CE Compile Job / FD-Clone-Linux (push) Has been cancelled
CE Compile Job / Show Code Archive Output (push) Has been cancelled
CE Compile Job / BUILD_SM8090 (push) Has been cancelled
CE Compile Job / BUILD_SM8689 (push) Has been cancelled
CE Compile Job / CE_UPLOAD (push) Has been cancelled
Deploy GitHub Pages / deploy (push) Has been cancelled
2025-08-28 23:46:25 +08:00
Yuanle Liu
4957908275 add input_processor plugin (#3657)
* add input_processor plugin

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update

* update
2025-08-28 22:53:57 +08:00
ming1753
02b3644903 [Bug Fix] VL Support w4a8/w4afp8 (#3686)
Some checks failed
CE Compile Job / ce_job_pre_check (push) Has been cancelled
CE Compile Job / print_ce_job_pre_check_outputs (push) Has been cancelled
CE Compile Job / FD-Clone-Linux (push) Has been cancelled
CE Compile Job / Show Code Archive Output (push) Has been cancelled
CE Compile Job / BUILD_SM8090 (push) Has been cancelled
CE Compile Job / BUILD_SM8689 (push) Has been cancelled
CE Compile Job / CE_UPLOAD (push) Has been cancelled
Deploy GitHub Pages / deploy (push) Has been cancelled
Publish Job / publish_pre_check (push) Has been cancelled
Publish Job / print_publish_pre_check_outputs (push) Has been cancelled
Publish Job / FD-Clone-Linux (push) Has been cancelled
Publish Job / Show Code Archive Output (push) Has been cancelled
Publish Job / BUILD_SM8090 (push) Has been cancelled
Publish Job / BUILD_SM8689 (push) Has been cancelled
Publish Job / PADDLE_PYPI_UPLOAD_8090 (push) Has been cancelled
Publish Job / PADDLE_PYPI_UPLOAD_8689 (push) Has been cancelled
Publish Job / Run FastDeploy Unit Tests and Coverage (push) Has been cancelled
Publish Job / Run FastDeploy LogProb Tests (push) Has been cancelled
Publish Job / Extracted partial CE model tasks to run in CI. (push) Has been cancelled
Publish Job / Run Base Tests (push) Has been cancelled
Publish Job / Run Accuracy Tests (push) Has been cancelled
2025-08-28 21:38:35 +08:00
YuanRisheng
808b548761 support tmp (#3675) 2025-08-28 19:42:32 +08:00
Divano
368bbd9dc6 Update _base_test.yml (#3690)
新增测试并发参数ci case
2025-08-28 19:15:19 +08:00
gaoziyuan
fc635acc47 [BugFix]fix dp&ep&tp and muti node infer (#3629)
* rm log

* fix bug

* fix bug

* fix dp&ep&tp and muti node infer

* fix

---------

Co-authored-by: Yuanle Liu <yuanlehome@163.com>
2025-08-28 19:09:10 +08:00
Divano
17731a8acd add concurrency cases (#3689) 2025-08-28 18:30:19 +08:00
Liumengyuan
2a73a6df03 fix_fp8_deepgemm_moe_tp_bug (#3658) 2025-08-28 17:19:02 +08:00
Liumengyuan
e93d4cfcdd Add with_output version AppendAttention (#3302)
* get use_output from fd_config

* add clear TODO description

* add mask_offset para to align with develop

* fix bug

* fix use_output logic

* fix sot bug
2025-08-28 17:10:18 +08:00
ltd0924
94ded434bd [BugFix] ep mixed offline exit (#3661)
* Update expert_service.py

* Update expert_service.py
2025-08-28 17:09:07 +08:00
ltd0924
e5015eea05 [BugFix] fix logger (#3666) 2025-08-28 17:08:00 +08:00
bukejiyu
73cf6096da fix (#3676)
* fix

* update
2025-08-28 17:06:32 +08:00
ltd0924
98c217b428 Update config.py (#3669) 2025-08-28 15:30:51 +08:00
co63oc
d4fc893fe3 fix typos (#3633)
Co-authored-by: Jiang-Jia-Jun <163579578+Jiang-Jia-Jun@users.noreply.github.com>
2025-08-28 14:42:24 +08:00
co63oc
c294fc8139 Fix target_version (#3159)
Some checks failed
CE Compile Job / ce_job_pre_check (push) Has been cancelled
CE Compile Job / print_ce_job_pre_check_outputs (push) Has been cancelled
CE Compile Job / FD-Clone-Linux (push) Has been cancelled
CE Compile Job / Show Code Archive Output (push) Has been cancelled
CE Compile Job / BUILD_SM8090 (push) Has been cancelled
CE Compile Job / BUILD_SM8689 (push) Has been cancelled
CE Compile Job / CE_UPLOAD (push) Has been cancelled
Deploy GitHub Pages / deploy (push) Has been cancelled
* Fix

* fix

* fix
2025-08-28 14:17:54 +08:00
Mattheliu
108d989d9d [Docs] add fastdeploy_unit_test_guide.md (#3484)
* docs:add fastdeploy_unit_test_guide.md

* docs:fix fastdeploy_unit_test_guide.md

* docs: add FastDeploy unit test spec (EN) and update usage nav

* fix codestyle
2025-08-28 14:12:25 +08:00
plusNew001
b791bea0c5 Update run_ci_xpu.sh to lock xvllm version (#3671)
Lock version due to xvllm update causing service errors.
2025-08-28 12:30:50 +08:00
Yuan Xiaolan
d37331fc71 fix w4afp8_gemm_scale_permute import error on A100 (#3611) 2025-08-28 11:42:23 +08:00
YuanRisheng
ad9b95e6dd fix rl bugs (#3654) 2025-08-28 11:09:34 +08:00
yangjianfengo1
e81046fdad 【New Feature】集中式支持w4afp8 (#3644)
* 支持tp w4afp8

* code style
2025-08-28 10:53:24 +08:00
周周周
76513f6416 Support 45t fp8 8 GPU (#3659) 2025-08-28 10:52:53 +08:00
Echo-Nie
7afcd4b776 【Hackathon 9th No.77】supplementary unit test for get_filtered_metrics (#3578)
* 功能模块 fastdeploy/metrics/metrics/get_filtered_metrics 单测补充

* fix pre-commit

---------

Co-authored-by: Tao Luo <luotao02@baidu.com>
2025-08-28 10:39:02 +08:00