zccjjj 20de04e249 [XPU] move xpu_attn_backend.py to FastDeploy/fastdeploy/model_executor/layers/backends/xpu (#5878)

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

..

benchmarks

…

cache_manager

[BugFix] fix dyc8 cache bug (#5958)

2026-01-08 19:25:47 -08:00

demo

…

distributed

…

engine

[Metax] fix shape error & output garbled code when reasoning big picture or video (#5965)

2026-01-09 13:41:45 +08:00

entrypoints

[KVCache] launch cache transfer processes only if hierarchical cache or kv cache storage is enabled (#5871)

2026-01-06 14:27:47 +08:00

eplb

…

golang_router

[Feature] Add Golang-based Router for Request Scheduling and Load Balancing (#5882)

2026-01-07 21:28:08 +08:00

input

[Models] Add Qwen3-VL Moe Model Support (#5913)

2026-01-08 11:36:42 +08:00

inter_communicator

[Feature] Support KV Cache Storage (#5571)

2025-12-25 16:30:35 +08:00

logger

…

metrics

…

model_executor

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

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

multimodal

[BugFix] fix dyc8 cache bug (#5958)

2026-01-08 19:25:47 -08:00

output

[Speculative Decoding] Return accepted tokens per head in response (#5947)

2026-01-09 14:32:08 +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

…

reasoning

…

rl

[Models] Add Qwen3-VL Moe Model Support (#5913)

2026-01-08 11:36:42 +08:00

router

…

scheduler

[BugFix] fix mm splitwise scheduler bug (#5604)

2025-12-25 04:08:11 -08:00

spec_decode

Revert "Revert "[TSP] last_norm allgather move to model.py (#5924)" (#5961)" (#5972)

2026-01-09 15:58:22 +08:00

splitwise

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

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

trace

set tracelogger stacklevel=2 (#5766)

2025-12-26 17:43:32 +08:00

transformer_utils

…

worker

[Speculative Decoding] Return accepted tokens per head in response (#5947)

2026-01-09 14:32:08 +08:00

__init__.py

[Feature] Add startup version check mechanism for Paddle (#5769)

2025-12-25 19:29:04 +08:00

collect_env.py

…

config.py

[Speculative Decoding] Optimize draft logprob (#5842)

2025-12-31 13:35:56 +08:00

envs.py

[Optimization] Reduce preemption occurrence when blocks not enough (#5696)

2026-01-07 20:01:16 +08:00

import_ops.py

…

stop.sh

…

test.yaml

…

utils.py

[BugFix] Fix redundant prompt_logprobs in the second chunk of streaming response when return_token_ids is enabled for v1/completions and fix trace file name (#5829)

2026-01-06 14:11:43 +08:00