apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-05 16:48:03 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[Bug fix] Fix prompt token ids dtype in v1 (#3860)

Browse Source
This commit is contained in:
chenjian
2025-09-08 11:34:13 +08:00
committed by GitHub
parent 7833f2f6cb
commit b5e20e3015
1 changed files with 4 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
fastdeploy/worker/gpu_model_runner.py
View File

@@ -290,10 +290,11 @@ class GPUModelRunner(ModelRunnerBase):
position_ids, request.get("max_tokens", 2048)
)
if len(request.output_token_ids) == 0:
input_ids = request.prompt_token_ids
if isinstance(request.prompt_token_ids, np.ndarray):
prompt_token_ids = request.prompt_token_ids.tolist()
else:
input_ids = request.prompt_token_ids + request.output_token_ids
prompt_token_ids = request.prompt_token_ids
input_ids = prompt_token_ids + request.output_token_ids
logger.debug(
f"Handle prefill request {request} at idx {idx}, "
f"{prefill_start_index=}, {prefill_end_index=}, "