chore: update whisper.cpp to v1.9.1 with RTX 50 (sm_120) CUDA support#4
Merged
gabrielste1n merged 1 commit intoJun 25, 2026
Merged
Conversation
gabrielste1n
approved these changes
Jun 25, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Brings the OpenWhispr whisper.cpp fork up to upstream v1.9.1 and extends the bundled CUDA builds to RTX 50-series (Blackwell,
sm_120) GPUs. The CUDA jobs inbuild-binaries.ymlpreviously targeted75;80;86;89on CUDA 12.4, which can't compile for Blackwell — nvcc 12.4 rejectscompute_120; this addssm_120and bumps the toolkit to 12.9.1 (Blackwell support landed in CUDA 12.8), keeping RTX 20-40 coverage. The produced CUDA binary was verified to embedsm_120SASS withcuobjdump --list-elf.The bulk of the diff is the upstream v1.8.3 → v1.9.1 update, squashed into a single commit; the OpenWhispr-specific change is
build-binaries.yml.Changes
.github/workflows/build-binaries.yml: addsm_120toCUDA_ARCHITECTURES(now75;80;86;89;120); bump CUDA toolkit 12.4.0 → 12.9.1 — Windows redist component versions and Linux package/paths (cuda-toolkit-12-9,/usr/local/cuda-12.9).