mirror of
https://github.com/PaddlePaddle/FastDeploy.git
synced 2025-12-24 13:28:13 +08:00
[Perf] Support tensor transmission between work and engine with zero-copy to improve efficiency (#4839)
* feat(zmq): support tensor transmission with zero-copy for improved efficiency * perf: zmq.send disable copy * zmq recv data for debug * convert logprobs tensor to cpu
This commit is contained in:
@@ -135,6 +135,7 @@ class SamplingParams:
|
||||
reasoning_max_tokens=None,
|
||||
min_tokens=1,
|
||||
logprobs=None,
|
||||
prompt_logprobs=None,
|
||||
bad_words=None,
|
||||
guided_decoding=None,
|
||||
bad_words_token_ids=None,
|
||||
@@ -158,6 +159,7 @@ class SamplingParams:
|
||||
reasoning_max_tokens=reasoning_max_tokens,
|
||||
min_tokens=min_tokens,
|
||||
logprobs=logprobs,
|
||||
prompt_logprobs=prompt_logprobs,
|
||||
bad_words=bad_words,
|
||||
guided_decoding=guided_decoding,
|
||||
bad_words_token_ids=bad_words_token_ids,
|
||||
|
||||
Reference in New Issue
Block a user