Skip to content
Test CUDA Builds

[ET-VK][matmul] Re-implement fp32/fp16 matmul and linear with tiled compute and blocked weight packing #11466

Sign in to view logs

[ET-VK][matmul] Re-implement fp32/fp16 matmul and linear with tiled compute and blocked weight packing

[ET-VK][matmul] Re-implement fp32/fp16 matmul and linear with tiled compute and blocked weight packing #11466

Run time

Learn about OS pricing on GitHub Actions
Job Run time
unittest-cuda / linux-job
27m 43s
test-executorch-cuda-build-12.6 / linux-job
19m 15s
test-executorch-cuda-build-12.8 / linux-job
19m 22s
test-executorch-cuda-build-13.0 / linux-job
18m 54s
test-models-cuda (resnet18) / linux-job
22m 39s
test-models-cuda (add) / linux-job
21m 9s
test-models-cuda (mv3) / linux-job
28m 38s
test-executorch-cuda-build-12.9 / linux-job
22m 1s
test-models-cuda (mv2) / linux-job
28m 9s
test-models-cuda (sdpa) / linux-job
21m 20s
test-models-cuda (conv1d) / linux-job
21m 16s
test-models-cuda (add_mul) / linux-job
21m 24s
export-model-cuda-artifact (Qwen, Qwen3-0.6B, quantized-int4-tile-packed) / linux-job
24m 39s
test-models-cuda (linear) / linux-job
21m 13s
export-model-cuda-artifact (mistralai, Voxtral-Mini-3B-2507, non-quantized) / linux-job
33m 45s
export-model-cuda-artifact (openai, whisper-large-v3-turbo, quantized-int4-tile-packed) / linux-job
26m 48s
export-model-cuda-artifact (openai, whisper-large-v3-turbo, quantized-int4-weight-only) / linux-job
25m 0s
export-model-cuda-artifact (nvidia, diar_streaming_sortformer_4spk-v2, non-quantized) / linux-job
33m 3s
export-model-cuda-artifact (nvidia, parakeet-tdt, quantized-int4-tile-packed) / linux-job
29m 26s
export-model-cuda-artifact (Qwen, Qwen3-0.6B, non-quantized) / linux-job
24m 37s
export-model-cuda-artifact (google, gemma-3-4b-it, non-quantized) / linux-job
36m 42s
export-model-cuda-artifact (nvidia, parakeet-tdt, non-quantized) / linux-job
28m 5s
export-model-cuda-artifact (openai, whisper-small, quantized-int4-tile-packed) / linux-job
25m 10s
export-model-cuda-artifact (Qwen, Qwen3-0.6B, quantized-int4-weight-only) / linux-job
24m 44s
export-model-cuda-artifact (mistralai, Voxtral-Mini-3B-2507, quantized-int4-tile-packed) / linux-job
35m 13s
export-model-cuda-artifact (nvidia, parakeet-tdt, quantized-int4-weight-only) / linux-job
28m 36s
export-model-cuda-artifact (openai, whisper-small, non-quantized) / linux-job
23m 16s
export-model-cuda-artifact (mistralai, Voxtral-Mini-4B-Realtime-2602, quantized-int4-tile-packed) / linux-job
42m 7s
export-model-cuda-artifact (openai, whisper-large-v3-turbo, non-quantized) / linux-job
24m 20s
export-model-cuda-artifact (google, gemma-3-4b-it, quantized-int4-tile-packed) / linux-job
35m 12s
export-model-cuda-artifact (openai, whisper-small, quantized-int4-weight-only) / linux-job
23m 29s
export-model-cuda-artifact (mistralai, Voxtral-Mini-3B-2507, quantized-int4-weight-only) / linux-job
34m 6s
check-all-cuda-builds
3s
test-cuda-pybind (qwen3-0.6b, Qwen-Qwen3-0.6B-cuda-non-quantized) / linux-job
22m 19s
test-cuda-pybind (gemma3-4b, --quantize, google-gemma-3-4b-it-cuda-quantized-int4-tile-packed) / linux-job
24m 27s
test-cuda-pybind (qwen3-0.6b, --quantize, Qwen-Qwen3-0.6B-cuda-quantized-int4-tile-packed) / linux-job
22m 10s
test-cuda-pybind (gemma3-4b, google-gemma-3-4b-it-cuda-non-quantized) / linux-job
25m 22s
test-model-cuda-e2e (openai, whisper-small, quantized-int4-weight-only) / linux-job
31m 23s
test-model-cuda-e2e (mistralai, Voxtral-Mini-3B-2507, quantized-int4-weight-only) / linux-job
17m 45s
test-model-cuda-e2e (google, gemma-3-4b-it, quantized-int4-tile-packed) / linux-job
17m 8s
test-model-cuda-e2e (google, gemma-3-4b-it, non-quantized) / linux-job
18m 19s
test-model-cuda-e2e (openai, whisper-small, non-quantized) / linux-job
29m 54s
test-model-cuda-e2e (openai, whisper-large-v3-turbo, non-quantized) / linux-job
31m 18s
test-model-cuda-e2e (openai, whisper-large-v3-turbo, quantized-int4-weight-only) / linux-job
29m 52s
test-model-cuda-e2e (nvidia, parakeet-tdt, quantized-int4-weight-only) / linux-job
16m 30s
test-model-cuda-e2e (openai, whisper-large-v3-turbo, quantized-int4-tile-packed) / linux-job
31m 8s
test-model-cuda-e2e (mistralai, Voxtral-Mini-4B-Realtime-2602, quantized-int4-tile-packed) / linux-job
35m 36s
test-model-cuda-e2e (mistralai, Voxtral-Mini-3B-2507, non-quantized) / linux-job
17m 51s
test-model-cuda-e2e (nvidia, parakeet-tdt, quantized-int4-tile-packed) / linux-job
16m 25s
test-model-cuda-e2e (nvidia, diar_streaming_sortformer_4spk-v2, non-quantized) / linux-job
16m 20s
test-model-cuda-e2e (mistralai, Voxtral-Mini-3B-2507, quantized-int4-tile-packed) / linux-job
16m 44s
test-model-cuda-e2e (openai, whisper-small, quantized-int4-tile-packed) / linux-job
30m 53s
test-model-cuda-e2e (nvidia, parakeet-tdt, non-quantized) / linux-job
16m 35s
21h 59m 23s