vllm
[Bugfix] Fix cuda graph sizes when running with speculative decoding
#30330
Merged
Overview
Commits
4
Changes
View On GitHub

[Bugfix] Fix cuda graph sizes when running with speculative decoding #30330

benchislett merged 4 commits into vllm-project:main from PatrykSaffer:patryk/cuda-graph-spec-dec-bug
PatrykSaffer
Patryk999 Fix cuda graph bug with spec dec
a954cb18
PatrykSaffer PatrykSaffer requested a review from WoosukKwon WoosukKwon 4 days ago
PatrykSaffer PatrykSaffer requested a review from youkaichao youkaichao 4 days ago
PatrykSaffer PatrykSaffer requested a review from robertgshaw2-redhat robertgshaw2-redhat 4 days ago
PatrykSaffer PatrykSaffer requested a review from mgoin mgoin 4 days ago
PatrykSaffer PatrykSaffer requested a review from tlrmchlsmth tlrmchlsmth 4 days ago
PatrykSaffer PatrykSaffer requested a review from houseroad houseroad 4 days ago
PatrykSaffer PatrykSaffer requested a review from hmellor hmellor 4 days ago
PatrykSaffer PatrykSaffer requested a review from yewentao256 yewentao256 4 days ago
PatrykSaffer PatrykSaffer requested a review from ProExpertProg ProExpertProg 4 days ago
chatgpt-codex-connector
mergify mergify added nvidia
gemini-code-assist
gemini-code-assist commented on 2025-12-09
PatrykSaffer Update vllm.py
a25c64e0
mergify
PatrykSaffer Update vllm.py
8edd14db
njhill njhill requested a review from benchislett benchislett 4 days ago
benchislett
benchislett approved these changes on 2025-12-09
benchislett benchislett added ready
benchislett Merge branch 'main' into patryk/cuda-graph-spec-dec-bug
c0f06863
benchislett benchislett enabled auto-merge (squash) 4 days ago
benchislett benchislett merged 4c2e10ea into main 4 days ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
ready nvidia
Milestone
No milestone
Terms Of Use Privacy Policy Refund Policy Impressum