Skip to content

Voxtral Realtime: enable streaming mode in CUDA CI #17844

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
mergennachin merged 1 commit into from
Mar 4, 2026
Merged

Voxtral Realtime: enable streaming mode in CUDA CI #17844

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
18 changes: 4 additions & 14 deletions .github/workflows/cuda.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -159,7 +159,7 @@ jobs:
repo: "google"
name: "gemma-3-4b-it"
quant: "quantized-int4-weight-only"
# Voxtral Realtime only supports int4-tile-packed on CUDA (offline mode)
# Voxtral Realtime only supports int4-tile-packed on CUDA
- model:
repo: "mistralai"
name: "Voxtral-Mini-4B-Realtime-2602"
Expand Down Expand Up @@ -197,12 +197,7 @@ jobs:
echo "::endgroup::"
fi

# Voxtral Realtime uses offline mode for CUDA CI (not streaming)
VR_MODE=""
if [ "${{ matrix.model.name }}" = "Voxtral-Mini-4B-Realtime-2602" ]; then
VR_MODE="vr-offline"
fi
source .ci/scripts/export_model_artifact.sh cuda "${{ matrix.model.repo }}/${{ matrix.model.name }}" "${{ matrix.quant }}" "${RUNNER_ARTIFACT_DIR}" "$VR_MODE"
source .ci/scripts/export_model_artifact.sh cuda "${{ matrix.model.repo }}/${{ matrix.model.name }}" "${{ matrix.quant }}" "${RUNNER_ARTIFACT_DIR}"

test-model-cuda-e2e:
name: test-model-cuda-e2e
Expand Down Expand Up @@ -237,7 +232,7 @@ jobs:
repo: "google"
name: "gemma-3-4b-it"
quant: "quantized-int4-weight-only"
# Voxtral Realtime only supports int4-tile-packed on CUDA (offline mode)
# Voxtral Realtime only supports int4-tile-packed on CUDA
- model:
repo: "mistralai"
name: "Voxtral-Mini-4B-Realtime-2602"
Expand All @@ -256,12 +251,7 @@ jobs:
download-artifact: ${{ matrix.model.repo }}-${{ matrix.model.name }}-cuda-${{ matrix.quant }}
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
script: |
# Voxtral Realtime uses offline mode for CUDA CI (not streaming)
VR_MODE=""
if [ "${{ matrix.model.name }}" = "Voxtral-Mini-4B-Realtime-2602" ]; then
VR_MODE="vr-offline"
fi
source .ci/scripts/test_model_e2e.sh cuda "${{ matrix.model.repo }}/${{ matrix.model.name }}" "${{ matrix.quant }}" "${RUNNER_ARTIFACT_DIR}" "$VR_MODE"
source .ci/scripts/test_model_e2e.sh cuda "${{ matrix.model.repo }}/${{ matrix.model.name }}" "${{ matrix.quant }}" "${RUNNER_ARTIFACT_DIR}"

test-cuda-pybind:
name: test-cuda-pybind
Expand Down
Loading