Skip to content

fix(qwen3-14b/prefill-tq): support long-context prompts via streaming attention (MAX_SEQ=4096) #773

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

Merged
zhangqi-chen merged 1 commit into from
Jul 23, 2026
Merged

fix(qwen3-14b/prefill-tq): support long-context prompts via streaming attention (MAX_SEQ=4096) #773

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
2 changes: 1 addition & 1 deletion models/qwen3/14b/qwen3_14b_prefill_tq_draft.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -74,7 +74,7 @@
# ---------------------------------------------------------------------------
# Prefill-specific tiling constants (local, may differ from config.py)
# ---------------------------------------------------------------------------
MAX_SEQ = 128
MAX_SEQ = M.max_seq # rope_cos/sin ctx cap; was 128 (draft placeholder) -> long prompts >128 read OOB rope -> garbage
BATCH_TILE = 16
K_CHUNK = 128
Q_OUT_CHUNK = 64
Expand Down
Loading