Ayakouji a8f8791668 [Optimization] Qwen2.5-VL support multi-batch prefill (#5269) ...

* update

* fix

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

* fix dict access

---------

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Co-authored-by: Jiang-Jia-Jun <163579578+Jiang-Jia-Jun@users.noreply.github.com>

2025-12-05 18:22:39 +08:00

..

benchmarks

[CLI]Update parameters in bench latecy cli tool and fix collect-env cli tool (#4558)

2025-10-24 16:46:45 +08:00

cache_manager

[PD Disaggregation] support DP via v1 router and decouple DP and EP (#5197)

2025-12-04 15:38:43 +08:00

demo

[PD Disaggregation] remove splitwise deployment on single node and refine the code (#4891)

2025-11-14 09:56:53 +08:00

distributed

fix import jit.marker.unified (#4622)

2025-10-28 22:11:03 +08:00

engine

[Optimization] Qwen2.5-VL support multi-batch prefill (#5269)

2025-12-05 18:22:39 +08:00

entrypoints

[Optimization] Qwen2.5-VL support multi-batch prefill (#5269)

2025-12-05 18:22:39 +08:00

eplb

[BugFix] fix cuda-python requirement (#5261)

2025-11-27 13:58:18 +08:00

input

[Bug fix] fix pooling models (#5358)

2025-12-04 11:06:30 +08:00

inter_communicator

[PD Disaggregation] support DP via v1 router and decouple DP and EP (#5197)

2025-12-04 15:38:43 +08:00

logger

[Log] Add trace log and add loggingInstrumentor tool (#4692)

2025-11-17 11:08:57 +08:00

metrics

[BugFix][Metrics] Fix Prometheus Multiprocess Metrics Issues and Add ZMQ Communication Metrics (#5185)

2025-11-27 15:05:09 +08:00

model_executor

[Feature] support Two batch overlap, mainly used in Prefill (#5078)

2025-12-05 14:58:50 +08:00

multimodal

[Bug Fix] process transparent image (#4807)

2025-11-05 17:15:24 +08:00

output

[Fearture] Support cache kv cache for output tokens (#4535)

2025-12-04 20:53:08 +08:00

platforms

[Feature] support flash_mask_attention backend (#5134)

2025-11-28 10:12:16 +08:00

plugins

[Feature] add cli command serve (#4226)

2025-09-24 14:50:45 +08:00

reasoning

[BugFix] Fix ernie_vl_reasoning_parsers.py 'end_token' to 'think_end_token' (#4805)

2025-11-06 11:28:55 +08:00

rl

[fix] update check_model_weights_status loop (#5249)

2025-12-04 19:43:01 +08:00

router

[PD Disaggregation] support DP via v1 router and decouple DP and EP (#5197)

2025-12-04 15:38:43 +08:00

scheduler

fix get_request from scheduler (#5369)

2025-12-04 21:59:10 +08:00

spec_decode

[Models] Add forward_meta to moe models' forward function (#5138)

2025-12-04 13:26:58 +08:00

splitwise

[PD Disaggregation] support DP via v1 router and decouple DP and EP (#5197)

2025-12-04 15:38:43 +08:00

trace

[Log] Add trace log and add loggingInstrumentor tool (#4692)

2025-11-17 11:08:57 +08:00

transformer_utils

[Feature] support pool (#3827)

2025-09-22 14:09:09 +08:00

worker

[Optimization] Qwen2.5-VL support multi-batch prefill (#5269)

2025-12-05 18:22:39 +08:00

__init__.py

remove fastsafetensors (#5371)

2025-12-04 19:22:04 +08:00

collect_env.py

feat: add support for API usage with multimodal models (#4548)

2025-10-28 20:23:46 +08:00

config.py

[Fearture] Support cache kv cache for output tokens (#4535)

2025-12-04 20:53:08 +08:00

envs.py

remove fastsafetensors (#5371)

2025-12-04 19:22:04 +08:00

import_ops.py

fix import error (#2944)

2025-07-22 14:06:01 +08:00

stop.sh

[PD Disaggregation] Support Qwen3-MoE use PD + EP inference. (#4691)

2025-11-06 10:32:15 +08:00

test.yaml

[Sync] Update to latest code (#2679)

2025-07-03 15:43:53 +08:00

utils.py

[LogProbs]Enable prompt logprobs output and modify data transmission method for the online interface. (#5089)

2025-12-02 13:49:51 +08:00