Skip to content

[Speculative Decoding]【Hackathon 10th Spring No.49】Adapt ngram_match and hybrid_mtp_ngram gpu kernels #7103

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
NKNaN wants to merge 2 commits into
base: develop
Choose a base branch
from
Open

[Speculative Decoding]【Hackathon 10th Spring No.49】Adapt ngram_match and hybrid_mtp_ngram gpu kernels #7103

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
6d1c5a0
adapting gpu ngram operator
NKNaN Mar 31, 2026
31fd29d
update
NKNaN Mar 31, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions custom_ops/gpu_ops/cpp_extensions.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -939,7 +939,9 @@ void NgramMatch(const paddle::Tensor& input_ids,
const paddle::Tensor& step_idx,
const paddle::Tensor& draft_token_num,
const paddle::Tensor& draft_tokens,
const paddle::Tensor& draft_tokens_copy,
const paddle::Tensor& seq_lens_this_time,
const paddle::Tensor& seq_lens_this_time_copy,
const paddle::Tensor& seq_lens_encoder,
const paddle::Tensor& seq_lens_decoder,
const paddle::Tensor& max_dec_len,
Expand All @@ -952,7 +954,9 @@ void HybridMtpNgram(const paddle::Tensor& input_ids,
const paddle::Tensor& step_idx,
const paddle::Tensor& draft_token_num,
const paddle::Tensor& draft_tokens,
const paddle::Tensor& draft_tokens_copy,
const paddle::Tensor& seq_lens_this_time,
const paddle::Tensor& seq_lens_this_time_copy,
const paddle::Tensor& seq_lens_decoder,
const paddle::Tensor& max_dec_len,
const int max_ngram_size,
Expand Down
Loading
Loading