OpenNMT / OpenNMT/CTranslate2

Feature request: CUDA 13 support

Open
#1,933 9 comments 43 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
C++
Stars
4.7k
Forks
536
Avg merge
12h 12m
Merged PRs (30d)
4

Description

Please add support for CUDA 13. I have a torch 2.9+cu13 stack with a blackwell dependency. While blackwell works with torch 2.7+ and can work with versions of CUDA supported by the current release here, I'm not in a position to downgrade from 13.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

No files, tests, or entry points are named in the issue. Start by reviewing the repository's existing CUDA support and compatibility checks, then trace how CUDA versions are selected and validated. Done means CTranslate2 supports the reported torch 2.9+cu13 and Blackwell stack without requiring a downgrade, with the relevant checks passing.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp, pytorch
Domain
build-system, machine-learning
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Active
Clarity
Mostly clear
Newbie friendliness
45/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.