Yonghua Li 833d00e2d7 [BugFix] move cache creation back to cache transfer process and adapt clear/update (#6144) ...

* [fix] move cache creation back to cache transfer process

* [fix] fix clear cache

* [chore] change some log level

* [fix] fix clear cache

* [fix] fix clear cache for blockwisefp8 and mtp

* [fix] fix c8

* [fix] fix clear_mtp_cache args

* [chore] update cache_transfer_manager

* [fix] fix update mtp cache

2026-01-24 21:59:13 +08:00

..

batch_invariant

…

benchmarks

[CI]Fix test cases failing under Python 3.12 (#6059)

2026-01-19 15:41:12 +08:00

cache_manager

[Feature] [KVCache] support attention_store kv cache backend (#5823)

2026-01-22 21:01:23 +08:00

ce

[BugFix] move cache creation back to cache transfer process and adapt clear/update (#6144)

2026-01-24 21:59:13 +08:00

ci_use

[BugFix] fix python3.12 v0_loader (#6132)

2026-01-21 16:12:11 +08:00

distributed

[Model Runner] Add exist_prefill_flag (#6172)

2026-01-23 13:07:05 +08:00

e2e

[UT] Add GLM E2E tests for non-MTP and MTP (#6163)

2026-01-23 10:34:29 +08:00

engine

[RL] add pause, update_weights, resume interface for async RL (#6052)

2026-01-23 10:18:07 +08:00

entrypoints

[RL] [APIServer] add more status codes for update/clear api (#6141)

2026-01-22 17:26:18 +08:00

eplb

[FDConfig] disable chunked_mm_input in ernie5 (#5774)

2025-12-26 15:31:27 +08:00

graph_optimization

[Graph Optimization] Add full_cuda_graph to control subgraph split (#6027)

2026-01-14 11:43:59 +08:00

input

【Optimization】update data_processor & add tool parser plugins (#6096)

2026-01-22 17:17:32 +08:00

inter_communicator

…

layers

[Feature] Unify fp8 block_wise quant ops (#5991)

2026-01-15 05:50:37 -08:00

logger

…

metax_ci

[Metax][CI] restore 'moe_expert_dispatch' outputs (#6130)

2026-01-21 10:33:09 +08:00

metrics

[log]console log to llm log (#5680)

2025-12-23 10:05:45 +08:00

model_executor

[Feature] Add PaddleFormers fallback backend (#5999)

2026-01-19 21:50:50 +08:00

model_loader

[Feature] Unify fp8 block_wise quant ops (#5991)

2026-01-15 05:50:37 -08:00

multimodal

…

operators

[BugFix] fix wint2 (#6109)

2026-01-20 21:46:21 +08:00

output

[Feature] Support stopping the inference for the corresponding request in the online service after a disconnection request. (#5320)

2026-01-16 11:46:13 +08:00

platforms

[XPU] move xpu_attn_backend.py to FastDeploy/fastdeploy/model_executor/layers/backends/xpu (#5878)

2026-01-09 16:34:57 +08:00

plugins

…

pooling

…

quantization

…

reasoning

…

rl

…

scheduler

[RL] add pause, update_weights, resume interface for async RL (#6052)

2026-01-23 10:18:07 +08:00

spec_decode

[Speculative Decoding] Support MTP for GLM-4.5-Air (#6047)

2026-01-16 14:35:24 +08:00

splitwise

[PD Disaggregation] Update usage of pd disaggregation and data parallel (#5742)

2026-01-05 17:51:29 +08:00

trace

…

usage

[CI]Fix test case (#6111)

2026-01-20 17:47:44 +08:00

utils

[Optimization] The pre- and post-processing pipeline do not perform dict conversion (#5494)

2026-01-22 00:50:52 +08:00

v1

[BugFix] fix mm revert bug (#6061)

2026-01-16 08:13:34 -08:00

worker

[Models] Add Qwen3-VL Model Support (#5763)

2025-12-29 17:39:33 +08:00

xpu_ci

[XPU] [CI] add xpu logprobs case (#6187)

2026-01-23 19:40:55 +08:00

conftest.py

…

cov_pytest.ini

[CI] Add 4-GPU e2e test job (#6082)

2026-01-19 10:42:14 +08:00

utils.py

[FDConfig] disable chunked_mm_input in ernie5 (#5774)

2025-12-26 15:31:27 +08:00