apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-07 01:22:59 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

Revert "[BugFix] Modify the bug in Qwen2 when enabling ENABLE_V1_KVCACHE_SCHE…" (#3719)

Browse Source 
This reverts commit 578b8c5da2.
This commit is contained in:
Jiang-Jia-Jun
2025-08-29 19:55:50 +08:00
committed by GitHub
parent 578b8c5da2
commit 3364af767b
2 changed files with 1 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
fastdeploy/engine/engine.py
View File

@@ -29,7 +29,6 @@ import traceback
import uuid
import weakref
from concurrent.futures import ThreadPoolExecutor
from dataclasses import asdict
from typing import Dict, List, Optional, Tuple
import numpy as np
@@ -480,8 +479,6 @@ class LLMEngine:
"""
# TODO 输入输出长度确认
if sampling_params is not None:
task.update(asdict(sampling_params))
request = Request.from_dict(task)
llm_logger.info(f"Receive request {request}")
if sampling_params is not None:

6
fastdeploy/worker/gpu_model_runner.py
View File

@@ -260,11 +260,7 @@ class GPUModelRunner(ModelRunnerBase):
self.share_inputs["need_think_end"][idx : idx + 1, :] = 0
self.share_inputs["reasoning_index"][idx : idx + 1, :] = 0
if len(request.output_token_ids) == 0:
input_ids = request.prompt_token_ids
else:
input_ids = request.prompt_token_ids + request.output_token_ids
input_ids = request.prompt_token_ids + request.output_token_ids
logger.debug(
f"Handle prefill request {request} at idx {idx} prefill_start_index {prefill_start_index} prefill_end_index {prefill_end_index} need_prefilled_token_num {len(input_ids)}"
)