apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-12-24 13:28:13 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[BugFix] fix thinking bug (#4710)

Browse Source 
* fix thinking bug

* fix ut

* update

* fix
This commit is contained in:
Yuanle Liu
2025-10-31 22:00:31 +08:00
committed by GitHub
parent 10358bf1a0
commit b301bd6c31
8 changed files with 458 additions and 290 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
custom_ops/gpu_ops/cpp_extensions.cc
View File

@@ -987,12 +987,15 @@ void LimitThinkingContentLengthV1(const paddle::Tensor& next_tokens,
const paddle::Tensor& max_think_lens,
const paddle::Tensor& step_idx,
const paddle::Tensor& limit_think_status,
const paddle::Tensor& stop_flags,
const paddle::Tensor& eos_token_ids,
const int64_t think_end_id);
void LimitThinkingContentLengthV2(const paddle::Tensor& next_tokens,
const paddle::Tensor& max_think_lens,
const paddle::Tensor& step_idx,
const paddle::Tensor& limit_think_status,
const paddle::Tensor& stop_flags,
const int64_t think_end_id,
const int64_t line_break_id);
@@ -1003,6 +1006,8 @@ void SpeculateLimitThinkingContentLengthV1(
const paddle::Tensor& limit_think_status,
const paddle::Tensor& accept_num,
const paddle::Tensor& seq_lens_decoder,
const paddle::Tensor& stop_flags,
const paddle::Tensor& eos_token_ids,
const int64_t think_end_id);
void SpeculateLimitThinkingContentLengthV2(
@@ -1012,6 +1017,7 @@ void SpeculateLimitThinkingContentLengthV2(
const paddle::Tensor& limit_think_status,
const paddle::Tensor& accept_num,
const paddle::Tensor& seq_lens_decoder,
const paddle::Tensor& stop_flags,
const int64_t think_end_id,
const int64_t line_break_id);