Skip to content
Test CUDA Builds

[cuda backend] skip fully-masked KV blocks calculation in SDPA #14048

Sign in to view logs

[cuda backend] skip fully-masked KV blocks calculation in SDPA

[cuda backend] skip fully-masked KV blocks calculation in SDPA #14048

Run time

Learn about OS pricing on GitHub Actions
Job Run time
Get changed files / get-changed-files
2s
CI run decision / decide
31s
unittest-cuda / linux-job
54m 13s
test-models-cuda / linux-job
48m 19s
test-executorch-cuda-build-12.6 / linux-job
20m 7s
test-executorch-cuda-build-13.0 / linux-job
18m 6s
export-model-cuda-artifact (Qwen, Qwen3-0.6B, quantized-int4-tile-packed) / linux-job
18m 4s
export-model-cuda-artifact (nvidia, parakeet-tdt, non-quantized) / linux-job
24m 37s
export-model-cuda-artifact (mistralai, Voxtral-Mini-4B-Realtime-2602, quantized-int4-tile-packed) / linux-job
23m 57s
export-model-cuda-artifact (mistralai, Voxtral-Mini-3B-2507, non-quantized) / linux-job
26m 44s
export-model-cuda-artifact (openai, whisper-large-v3-turbo, quantized-int4-tile-packed) / linux-job
20m 45s
export-model-cuda-artifact (openai, whisper-large-v3-turbo, quantized-int4-weight-only) / linux-job
18m 53s
export-model-cuda-artifact (google, gemma-3-4b-it, quantized-int4-tile-packed) / linux-job
29m 6s
export-model-cuda-artifact (nvidia, parakeet-tdt, quantized-int4-tile-packed) / linux-job
26m 20s
export-model-cuda-artifact (mistralai, Voxtral-Mini-3B-2507, quantized-int4-tile-packed) / linux-job
25m 11s
export-model-cuda-artifact (SocialLocalMobile, gemma-4-31B-it-HQQ-INT4, quantized-int4-tile-packed) / linux-job
50m 33s
export-model-cuda-artifact (SocialLocalMobile, Qwen3.5-35B-A3B-HQQ-INT4, quantized-int4-tile-packed) / linux-job
43m 4s
export-model-cuda-artifact (facebook, dinov2-small-imagenet1k-1-layer, non-quantized) / linux-job
13m 45s
export-model-cuda-artifact (nvidia, diar_streaming_sortformer_4spk-v2, non-quantized) / linux-job
31m 30s
export-model-cuda-artifact (openai, whisper-small, non-quantized) / linux-job
17m 23s
export-model-cuda-artifact (mistralai, Voxtral-Mini-3B-2507, quantized-int4-weight-only) / linux-job
27m 14s
export-model-cuda-artifact (Qwen, Qwen3-0.6B, non-quantized) / linux-job
17m 2s
export-model-cuda-artifact (google, gemma-3-4b-it, non-quantized) / linux-job
29m 3s
check-all-cuda-builds
3s
test-cuda-pybind (gemma3-4b, --quantize, google-gemma-3-4b-it-cuda-quantized-int4-tile-packed) / linux-job
22m 38s
test-cuda-pybind (qwen3-0.6b, Qwen-Qwen3-0.6B-cuda-non-quantized) / linux-job
20m 57s
test-cuda-pybind (qwen3-0.6b, --quantize, Qwen-Qwen3-0.6B-cuda-quantized-int4-tile-packed) / linux-job
20m 31s
test-model-cuda-e2e (openai, whisper-large-v3-turbo, quantized-int4-weight-only) / linux-job
24m 41s
test-model-cuda-e2e (mistralai, Voxtral-Mini-4B-Realtime-2602, quantized-int4-tile-packed) / linux-job
26m 33s
test-model-cuda-e2e (nvidia, parakeet-tdt, non-quantized) / linux-job
10m 9s
test-model-cuda-e2e (mistralai, Voxtral-Mini-3B-2507, non-quantized) / linux-job
10m 37s
test-model-cuda-e2e (nvidia, diar_streaming_sortformer_4spk-v2, non-quantized) / linux-job
9m 36s
test-model-cuda-e2e (mistralai, Voxtral-Mini-3B-2507, quantized-int4-weight-only) / linux-job
11m 0s
test-model-cuda-e2e (openai, whisper-small, non-quantized) / linux-job
25m 6s
test-model-cuda-e2e (SocialLocalMobile, gemma-4-31B-it-HQQ-INT4, quantized-int4-tile-packed) / linux-job
15m 8s
test-model-cuda-e2e (SocialLocalMobile, Qwen3.5-35B-A3B-HQQ-INT4, quantized-int4-tile-packed) / linux-job
15m 34s
test-model-cuda-e2e (google, gemma-3-4b-it, quantized-int4-tile-packed) / linux-job
10m 7s
test-model-cuda-e2e (openai, whisper-large-v3-turbo, quantized-int4-tile-packed) / linux-job
25m 11s
test-model-cuda-e2e (facebook, dinov2-small-imagenet1k-1-layer, non-quantized) / linux-job
8m 11s
test-model-cuda-e2e (google, gemma-3-4b-it, non-quantized) / linux-job
12m 3s
test-model-cuda-e2e (nvidia, parakeet-tdt, quantized-int4-tile-packed) / linux-job
8m 52s
test-model-cuda-e2e (mistralai, Voxtral-Mini-3B-2507, quantized-int4-tile-packed) / linux-job
10m 9s
14h 31m 35s