kxz2002
cbe27ad9fb
[Cherry-Pick] Fix ernie_vl_reasoning_parsers.py 'end_token' to 'think_end_token' ( #4805 ) ( #4842 )
...
* [Feature] add a new reasoning parser (#4571 )
* add new reasoning_parser initial commit
* add parser file content
* add register
* ernie_test_reasoning_parser
* support <tool_call> token and add tool_parser
* add and fix unit tests
* modify reasoning_parser
* modify reasoning parser and tool parser
* modify unit tests
* modify reasoning_parser and tool_parser
* modify unit tests
* fix tool_parser
* modify the logic of reasoning_parser and tool_parser
* add and modify unit tests
* standardize code style
* simplify reasoning_parser and tool_parser
* modify unit test
* [BugFix] Fix finish reason in _create_chat_completion_choice (#4582 )
* fix n_param _create_chat_completion_choicel
* fix unit test
* fix final_res
* modify unit tests
* [BugFix] fix offline llm chat "enable_thinking" is always "False" (#4686 )
* fix enable_thinking
* recover ernie4_5_vl_processor
* [BugFix] Fix ernie_vl_reasoning_parsers.py 'end_token' to 'think_end_token' (#4805 )
* fix ernie_vl_reasoning_parsers.py 'end_token' to 'think_end_token'
* add unit tests
2025-11-06 15:54:48 +08:00
ApplEOFDiscord
8e48da8027
[Bug Fix] process transparent image ( #4807 ) ( #4847 )
...
* process transparent image
* english comments
* process transparency at downloading
* fix
* remove useless codes
2025-11-06 15:43:44 +08:00
RAM
93aedaf23e
temporary fix bug of 0 size tensor ( #4844 )
2025-11-06 14:20:59 +08:00
RichardWooSJTU
eac823aec8
[BugFix] fix total_block_num init error in worker_process ( #4553 ) ( #4817 )
...
* fix total_block_num init error in worker_process
* fix req and token client
* fix req and token client
* fix xpu xi
* fix xpu ci
2025-11-06 13:59:57 +08:00
yinwei
0df488c7bb
support wint8 & wint4 ( #4837 )
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
2025-11-06 10:54:34 +08:00
kxz2002
a218291831
[Cherry-Pick] Fix reasoning parser register name ( #4795 ) ( #4816 )
...
* [Feature] add a new reasoning parser (#4571 )
* add new reasoning_parser initial commit
* add parser file content
* add register
* ernie_test_reasoning_parser
* support <tool_call> token and add tool_parser
* add and fix unit tests
* modify reasoning_parser
* modify reasoning parser and tool parser
* modify unit tests
* modify reasoning_parser and tool_parser
* modify unit tests
* fix tool_parser
* modify the logic of reasoning_parser and tool_parser
* add and modify unit tests
* standardize code style
* simplify reasoning_parser and tool_parser
* modify unit test
* [BugFix] Fix finish reason in _create_chat_completion_choice (#4582 )
* fix n_param _create_chat_completion_choicel
* fix unit test
* fix final_res
* modify unit tests
* [BugFix] fix offline llm chat "enable_thinking" is always "False" (#4686 )
* fix enable_thinking
* recover ernie4_5_vl_processor
* fix parser register name (#4795 )
Co-authored-by: luukunn <83932082+luukunn@users.noreply.github.com >
---------
Co-authored-by: luukunn <83932082+luukunn@users.noreply.github.com >
2025-11-06 10:51:40 +08:00
李泳桦
7e7a91855b
[BugFix] fix messages being inplace modified in offline chat api ( #4830 )
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
2025-11-05 20:46:55 +08:00
kxz2002
1689f7ef86
[Cherry-Pick] Fix ernie4_5_vl_processor.py and qwen_vl_processor.py can not disable thinking ( #4762 ) ( #4798 )
...
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
* [Feature] add a new reasoning parser (#4571 )
* add new reasoning_parser initial commit
* add parser file content
* add register
* ernie_test_reasoning_parser
* support <tool_call> token and add tool_parser
* add and fix unit tests
* modify reasoning_parser
* modify reasoning parser and tool parser
* modify unit tests
* modify reasoning_parser and tool_parser
* modify unit tests
* fix tool_parser
* modify the logic of reasoning_parser and tool_parser
* add and modify unit tests
* standardize code style
* simplify reasoning_parser and tool_parser
* modify unit test
* [BugFix] Fix finish reason in _create_chat_completion_choice (#4582 )
* fix n_param _create_chat_completion_choicel
* fix unit test
* fix final_res
* modify unit tests
* [BugFix] fix offline llm chat "enable_thinking" is always "False" (#4686 )
* fix enable_thinking
* recover ernie4_5_vl_processor
* [BugFix] Fix ernie4_5_vl_processor.py and qwen_vl_processor.py can not disable thinking (#4762 )
* fix ernie4_5_vl_processor.py and qwen_vl_processor.py
* add unit test
2025-11-05 10:59:47 +08:00
YuBaoku
7cee8030af
[CI] Disable unstable test jobs and cases ( #4799 )
...
[CI] Disable unstable test jobs and cases
2025-11-05 10:28:53 +08:00
李泳桦
61856e55ce
[fix] fix v0 pd, let worker step_shm_value create=False ( #4781 )
...
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
Co-authored-by: xiegegege <46314656+xiegegege@users.noreply.github.com >
2025-11-04 20:38:01 +08:00
ming1753
08abb0dd1c
[Docs] Add new support models ( #4800 )
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
2025-11-04 16:49:40 +08:00
Lucas
6502d19e97
[XPU] add deploy doc for PaddleOCR-VL in XPU ( #4792 )
2025-11-04 15:07:04 +08:00
ddchenhao66
78a1451eb7
[XPU] fix thinking bug where output only contains reasoning_content ( #4760 )
...
Co-authored-by: ddchenhao66 <dhaochen163.com>
2025-11-04 12:47:34 +08:00
plusNew001
ffa57dbfac
Modify base_response_104 for better clarity ( #4789 )
...
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
Updated the base_response_104 message to improve wording.
2025-11-03 22:08:32 +08:00
kevin
cf5e545a73
test scheduler ( #4757 )
2025-11-03 20:12:02 +08:00
ming1753
e5cbe0d6a1
[Docs] PaddleOCR-VL add RTX3060 server param ( #4766 )
...
* [Docs] PaddleOCR-VL add RTX3060 server param
* modify config
* fix bug
2025-11-03 19:54:59 +08:00
luukunn
7df7035055
【DataProcessor】add options thinking_mode ( #4735 ) ( #4759 )
...
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
* add thinking_mode
* add thinking_mode
* add thinking_mode
* add thinking_mode
* add thinking_mode
* add thinking_mode
* add unit test
2025-11-03 18:14:39 +08:00
yyssys
6bfa4fed6e
[Docs]Update XPU document version to 2.3.0 ( #4754 )
...
* [Docs]Update XPU document version to 2.3.0
* update
2025-11-03 13:36:12 +08:00
Yuanle Liu
4d4c13f1b5
[CherryPick] Fix thinking bug cp ( #4736 )
...
* fix thinking bug
* fix ut
* update
* fix
v2.3.0-rc0
2025-11-03 11:41:03 +08:00
Jundong Liu
d1d3876c16
[FDConfig] [PD Disaggregation] [Graph Optimization] Close Cudagraph for P node when PD Disaggregation ( #4632 ) ( #4734 )
...
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
* Close cudagraph for P node when PD Disaggregation
* fix problem
2025-11-03 10:59:34 +08:00
YuBaoku
8632b778f5
[CI] update paddlepaddle_gpu==3.2.1 and fix rollout_model test logic ( #4738 )
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
2025-11-02 21:30:23 +08:00
yinwei
b4aa189483
[XPU] Support V1 Loader in Bf16 ( #4746 )
...
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
* add v1 support for bf16
* update
* update
* update
* update
* update
* update code
2025-11-01 16:13:25 +08:00
kxz2002
0861eb027f
[Cherry-Pick] Fix finish reason in _create_chat_completion_choice ( #4582 ) ( #4716 )
...
* [Feature] add a new reasoning parser (#4571 )
* add new reasoning_parser initial commit
* add parser file content
* add register
* ernie_test_reasoning_parser
* support <tool_call> token and add tool_parser
* add and fix unit tests
* modify reasoning_parser
* modify reasoning parser and tool parser
* modify unit tests
* modify reasoning_parser and tool_parser
* modify unit tests
* fix tool_parser
* modify the logic of reasoning_parser and tool_parser
* add and modify unit tests
* standardize code style
* simplify reasoning_parser and tool_parser
* modify unit test
* [BugFix] Fix finish reason in _create_chat_completion_choice (#4582 )
* fix n_param _create_chat_completion_choicel
* fix unit test
* fix final_res
* modify unit tests
* [BugFix] fix offline llm chat "enable_thinking" is always "False" (#4686 )
* fix enable_thinking
* recover ernie4_5_vl_processor
2025-11-01 15:50:47 +08:00
kxz2002
24b85b752b
[Cherry-Pick] Unify the registration name recognition for tool_parser and reasoning_parser to “-” ( #4668 ) ( #4737 )
...
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
* [Feature] add a new reasoning parser (#4571 )
* add new reasoning_parser initial commit
* add parser file content
* add register
* ernie_test_reasoning_parser
* support <tool_call> token and add tool_parser
* add and fix unit tests
* modify reasoning_parser
* modify reasoning parser and tool parser
* modify unit tests
* modify reasoning_parser and tool_parser
* modify unit tests
* fix tool_parser
* modify the logic of reasoning_parser and tool_parser
* add and modify unit tests
* standardize code style
* simplify reasoning_parser and tool_parser
* modify unit test
* [BugFix] Fix finish reason in _create_chat_completion_choice (#4582 )
* fix n_param _create_chat_completion_choicel
* fix unit test
* fix final_res
* modify unit tests
* [BugFix] fix offline llm chat "enable_thinking" is always "False" (#4686 )
* fix enable_thinking
* recover ernie4_5_vl_processor
* [Feature] Unify the registration name recognition for tool_parser and reasoning_parser to “-” (#4668 )
* parser register name unify
* change ernie_x1 to ernie-x1
* change ernie4_5_vl to ernie-45-vl
* fix unit test
2025-10-31 23:27:21 +08:00
Longzhi Wang
d11e27a188
[Bugfix] fix test_get_save_output_v1 ( #4732 )
2025-10-31 22:52:04 +08:00
ming1753
96a44c8574
Skip building native architecture when specifying arch list ( #4728 )
2025-10-31 20:45:49 +08:00
plusNew001
e4463c37fe
[XPU][CI] Release ci fix bug ( #4742 )
...
* Clean up XVLLM_PATH setup in run_ci_xpu.sh
Removed XVLLM_PATH setup and related wget command.
* Update xvllm version in download_dependencies.sh
* Update run_ci_xpu.sh
2025-10-31 20:30:35 +08:00
ddchenhao66
ce53cdccd2
[XPU] xpu support neox style ROPE ( #4723 )
...
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
Co-authored-by: ddchenhao66 <dhaochen163.com>
2025-10-31 18:17:21 +08:00
RAM
00d0da0c18
[Graph Optimization] Add the CUDAGraph usage switch for Draft Model ( #4669 )
...
* add draft model using cudagraph switch
* set default as false
* capture draft model in ci
---------
Co-authored-by: Jiang-Jia-Jun <163579578+Jiang-Jia-Jun@users.noreply.github.com >
2025-10-31 17:34:09 +08:00
Ryan
9a647cb61c
[OP] Add InferShape&InferDtype for per_token_quant_padding ( #4667 ) ( #4683 )
...
* add InferShape&InferDtype for per_token_quant_padding
* fix codestyle
2025-10-31 16:42:52 +08:00
plusNew001
7b013c63e2
[XPU ][CI] Release XPU ci update ( #4722 )
...
* Refactor CI script for paddlepaddle-xpu installation
Updated the CI script to install specific paddlepaddle-xpu version and modified the testing commands for better performance and error handling.
* Add test script for OpenAI client interaction
* Remove empty line at the beginning of run_45vl.py
2025-10-31 15:36:14 +08:00
kevin
139342d953
fix bug ( #4680 )
2025-10-31 15:23:33 +08:00
李泳桦
9cf4005e62
[Cherry-pick] Fix profile run in pd-disaggregated deployment ( #4693 )
...
* [fix] fix pd+dp+ep bug
* [fix] fix again
* [ci] fix code style
2025-10-31 14:41:35 +08:00
chen
802dfa6524
fix --logprobs-mode raw_logits ( #4681 ) ( #4712 )
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
2025-10-31 10:50:31 +08:00
ddchenhao66
2e7b7a42c2
[XPU] xpu currently disable prefix cache for VL model ( #4694 )
...
Co-authored-by: ddchenhao66 <dhaochen163.com>
2025-10-31 10:37:41 +08:00
ming1753
40b87065cc
fix docs bug ( #4703 )
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
2025-10-30 19:53:40 +08:00
ming1753
9defdaed6b
[BugFix] Fix PaddleOCRVL bug ( #4678 )
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
2025-10-30 13:49:08 +08:00
ApplEOFDiscord
52a6e0be41
[Cherry-Pick] add mm token usage ( #4648 )
...
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
* [Feature] add mm token usage (#4570 )
* add mm token usage
* fix unit test
* fix unit test
* fix unit test
* fix model path
* fix unit test
* fix unit test
* fix unit test
* remove uncomment
* change var name
* fix code style
* fix code style
* fix code style
* fix code style
* fix unit test
* update doc
* update doc
---------
Co-authored-by: Jiang-Jia-Jun <163579578+Jiang-Jia-Jun@users.noreply.github.com >
2025-10-30 09:58:07 +08:00
kxz2002
895ca7694e
[Feature] add a new reasoning parser ( #4571 ) ( #4664 )
...
* add new reasoning_parser initial commit
* add parser file content
* add register
* ernie_test_reasoning_parser
* support <tool_call> token and add tool_parser
* add and fix unit tests
* modify reasoning_parser
* modify reasoning parser and tool parser
* modify unit tests
* modify reasoning_parser and tool_parser
* modify unit tests
* fix tool_parser
* modify the logic of reasoning_parser and tool_parser
* add and modify unit tests
* standardize code style
* simplify reasoning_parser and tool_parser
* modify unit test
2025-10-30 09:49:53 +08:00
Lucas
df72033adb
[XPU] fix pos_emb_type bug ( #4639 )
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
2025-10-29 17:14:47 +08:00
ming1753
7b275efc59
[Docs] Add PaddleOCR-VL-0.9B best practices ( #4661 )
2025-10-29 16:58:38 +08:00
ddchenhao66
21bb2d69d1
[XPU] Update the return value of TextImageGatherScatter ( #4646 )
...
Co-authored-by: ddchenhao66 <dhaochen163.com>
2025-10-29 16:17:23 +08:00
RAM
a0d5426ab6
fix ci test case ( #4635 )
2025-10-29 13:26:38 +08:00
xiaolei373
14e7d88ea4
[feature] support reward api ( #4518 )
...
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
Co-authored-by: SunLei <sunlei5788@gmail.com >
2025-10-29 00:20:28 +08:00
李泳桦
a012e3608b
[Feature] support logits processors ( #4515 )
...
* [feat] provide an interface for logits processors and a builtin LogitBiasLogitsProcessor
* [chore] fix code style
* [fix] add unit test & fix existing bugs
* [feat] add engine/worker arg --logits-processors
* [fix] redefine user args as logits_processors_args and fix some bugs
* [fix] fix test_sampler
* Update fastdeploy/model_executor/logits_processor/builtin.py
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com >
* Update fastdeploy/model_executor/logits_processor/__init__.py
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com >
* Update tests/model_executor/test_logits_processor.py
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com >
* [fix] fix typo
* Update fastdeploy/engine/sampling_params.py
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com >
* [fix] fix bracelet
* [chore] redefine logits processor interface: pass the entire share_inputs into LP, do not copy share_inputs and logits
* [doc] add docs
* [fix] fix logit bias processor not applied when decoding is too fast & add docs and tests
* [fix] fix redundant code
* [feat] skip apply() if no bias is specified
---------
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com >
2025-10-29 00:08:53 +08:00
zhang-prog
24b9505971
add einops dependency ( #4633 )
2025-10-28 22:17:13 +08:00
Yuanle Liu
20756cd2bb
fix import jit.marker.unified ( #4622 )
2025-10-28 22:11:03 +08:00
ming1753
561b9f38d3
[BugFix] fix paddleocr prefix cache bug ( #4625 )
...
* fix paddleocr prefix cache bug
* disable prefix-caching in ocr
2025-10-28 21:38:12 +08:00
RAM
fff5fb5e39
[Graph Optimization] Refactor default capture list ( #4617 )
...
* fix bug and refine code
* add debug count
* refine code
2025-10-28 21:31:02 +08:00
Lucas
0a0c74e717
[XPU] Support PaddleOCR-VL model for XPU ( #4529 )
...
* [XPU] support PaddleOCR-VL in XPU
* [XPU] fix PaddleOCR-VL pos_emb_type
2025-10-28 20:35:04 +08:00