Skip to content
trunk

Add W4A8 INT8 activation kernels for batched MoE prefill #24209

Sign in to view logs

Add W4A8 INT8 activation kernels for batched MoE prefill

Add W4A8 INT8 activation kernels for batched MoE prefill #24209

Run time

Learn about OS pricing on GitHub Actions
Job Run time
changes
7s
test-static-llama-ane / macos-job
9m 40s
test-llama-torchao-lowbit / macos-job
11m 23s
test-custom-ops-macos (cmake) / macos-job
14m 55s
test-coreml-delegate / macos-job
1m 29s
test-selective-build-macos (cmake) / macos-job
16m 38s
test-demo-backend-delegation (cmake) / linux-job
8m 44s
test-arm-backend-zephyr (ethos-u55) / linux-job
16m 45s
test-arm-backend-zephyr (ethos-u85) / linux-job
17m 15s
test-demo-backend-delegation (buck2) / linux-job
9m 17s
test-arm-backend-zephyr (cortex-m55) / linux-job
17m 26s
test-models-macos-cpu (vit, xnnpack-quantization-delegation) / macos-job
14m 30s
test-models-macos-cpu (efficient_sam, portable) / macos-job
26m 57s
test-models-macos-cpu (mv3, portable) / macos-job
14m 22s
test-models-macos-cpu (llama3_2_vision_encoder, portable) / macos-job
22m 54s
test-models-macos-cpu (w2l, xnnpack-quantization-delegation) / macos-job
13m 45s
test-models-macos-cpu (llama, portable) / macos-job
14m 10s
test-models-macos-cpu (emformer_join, xnnpack-quantization-delegation) / macos-job
13m 45s
test-models-macos-cpu (ic4, xnnpack-quantization-delegation) / macos-job
14m 9s
test-models-macos-cpu (resnet50, xnnpack-quantization-delegation) / macos-job
13m 59s
test-models-macos-cpu (mv3, xnnpack-quantization-delegation) / macos-job
14m 2s
test-models-macos-cpu (llama2, xnnpack-quantization-delegation) / macos-job
13m 54s
test-models-linux-aarch64 (mobilebert, portable, linux.arm64.2xlarge) / linux-job
17m 9s
test-models-linux-aarch64 (add_mul, xnnpack-quantization-delegation, linux.arm64.2xlarge) / linux-job
11m 29s
test-models-linux-aarch64 (emformer_join, xnnpack-quantization-delegation, linux.arm64.2xlarge) / linux-job
11m 26s
test-models-macos-cpu (mobilebert, xnnpack-quantization-delegation) / macos-job
15m 16s
test-models-linux-aarch64 (llama3_2_vision_encoder, portable, linux.arm64.2xlarge) / linux-job
44m 17s
test-models-linux-aarch64 (resnet50, portable, linux.arm64.2xlarge) / linux-job
30m 53s
test-models-linux-aarch64 (vit, portable, linux.arm64.2xlarge) / linux-job
20m 28s
test-models-linux-aarch64 (emformer_transcribe, xnnpack-quantization-delegation, linux.arm64.2xla... / linux-job
19m 56s
test-models-linux-aarch64 (mul, portable, linux.arm64.2xlarge) / linux-job
11m 16s
test-models-linux-aarch64 (add_mul, portable, linux.arm64.2xlarge) / linux-job
11m 19s
test-models-linux-aarch64 (mv3, xnnpack-quantization-delegation, linux.arm64.2xlarge) / linux-job
12m 16s
test-models-linux-aarch64 (resnet18, xnnpack-quantization-delegation, linux.arm64.2xlarge) / linux-job
11m 56s
test-models-linux-aarch64 (ic4, xnnpack-quantization-delegation, linux.arm64.2xlarge) / linux-job
13m 53s
test-models-linux-aarch64 (mv2, xnnpack-quantization-delegation, linux.arm64.2xlarge) / linux-job
12m 2s
test-models-linux-aarch64 (add, xnnpack-quantization-delegation, linux.arm64.2xlarge) / linux-job
11m 40s
test-models-linux-aarch64 (emformer_join, portable, linux.arm64.2xlarge) / linux-job
45m 48s
test-models-linux-aarch64 (softmax, portable, linux.arm64.2xlarge) / linux-job
11m 19s
test-models-linux-aarch64 (mv2, portable, linux.arm64.2xlarge) / linux-job
13m 12s
test-models-linux-aarch64 (w2l, portable, linux.arm64.2xlarge) / linux-job
13m 20s
test-models-linux-aarch64 (qwen2_5_1_5b, portable, linux.arm64.2xlarge) / linux-job
16m 42s
test-models-linux-aarch64 (linear, xnnpack-quantization-delegation, linux.arm64.2xlarge) / linux-job
11m 19s
test-models-linux-aarch64 (mv3, portable, linux.arm64.2xlarge) / linux-job
11m 58s
test-models-linux-aarch64 (add, portable, linux.arm64.2xlarge) / linux-job
11m 18s
test-models-linux-aarch64 (ic3, portable, linux.arm64.2xlarge) / linux-job
23m 59s
test-models-linux-aarch64 (lstm, portable, linux.arm64.2xlarge) / linux-job
11m 25s
test-models-linux-aarch64 (ic4, portable, linux.arm64.2xlarge) / linux-job
1h 19m 47s
test-models-linux-aarch64 (ic3, xnnpack-quantization-delegation, linux.arm64.2xlarge) / linux-job
12m 39s
test-models-linux-aarch64 (emformer_transcribe, portable, linux.arm64.2xlarge) / linux-job
13m 0s
test-models-linux-aarch64 (resnet50, xnnpack-quantization-delegation, linux.arm64.2xlarge) / linux-job
12m 8s
test-models-linux-aarch64 (vit, xnnpack-quantization-delegation, linux.arm64.2xlarge) / linux-job
13m 20s
test-models-linux-aarch64 (mobilebert, xnnpack-quantization-delegation, linux.arm64.2xlarge) / linux-job
18m 45s
test-models-linux-aarch64 (phi_4_mini, portable, linux.arm64.m7g.4xlarge) / linux-job
59m 38s
test-models-linux-aarch64 (linear, portable, linux.arm64.2xlarge) / linux-job
11m 10s
test-models-linux-aarch64 (w2l, xnnpack-quantization-delegation, linux.arm64.2xlarge) / linux-job
11m 19s
test-huggingface-transformers-xnnpack (llama3.2-1b|xnnpack|--quantize) / linux-job
20m 29s
test-models-linux-aarch64 (resnet18, portable, linux.arm64.2xlarge) / linux-job
17m 48s
test-huggingface-transformers-xnnpack (smollm3-3b|xnnpack|--quantize) / linux-job
32m 36s
test-huggingface-transformers-xnnpack (qwen3-0.6b|xnnpack|--quantize) / linux-job
23m 7s
test-huggingface-transformers-xnnpack (gemma3-1b|xnnpack|--quantize) / linux-job
26m 37s
test-huggingface-transformers-xnnpack (smollm2-135m|xnnpack|--quantize) / linux-job
20m 40s
test-torchao-huggingface-checkpoints (qwen3_4b, linux.arm64.2xlarge, executorch-ubuntu-22.04-gcc1... / linux-job
40m 50s
test-huggingface-transformers-xnnpack (qwen3-1.7b|xnnpack|--quantize) / linux-job
40m 2s
test-torchao-huggingface-checkpoints (lfm2_5_1_2b, linux.2xlarge, executorch-ubuntu-22.04-clang12... / linux-job
17m 11s
test-torchao-huggingface-checkpoints (phi_4_mini, linux.arm64.2xlarge, executorch-ubuntu-22.04-gc... / linux-job
27m 56s
test-torchao-huggingface-checkpoints (lfm2_5_1_2b, linux.arm64.2xlarge, executorch-ubuntu-22.04-g... / linux-job
14m 45s
test-torchao-huggingface-checkpoints (phi_4_mini, linux.2xlarge, executorch-ubuntu-22.04-clang12,... / linux-job
29m 30s
test-torchao-huggingface-checkpoints (qwen3_4b, linux.2xlarge, executorch-ubuntu-22.04-clang12, x... / linux-job
37m 9s
test-huggingface-transformers-macos (qwen3-0.6b|coreml_fp32_gpu|--quantize) / macos-job
10m 45s
test-huggingface-transformers-macos (olmo-1b|coreml_fp32_gpu|--quantize) / macos-job
11m 43s
test-huggingface-transformers-macos (smollm2-135m|coreml_fp32_gpu|--quantize) / macos-job
10m 25s
test-huggingface-transformers-macos (llama3.2-1b|coreml_fp32_gpu|--quantize) / macos-job
10m 56s
test-huggingface-transformers-macos (distilbert|coreml_fp32_gpu|--quantize) / macos-job
9m 39s
test-qnn-optimum-model (fp32, pvt) / linux-job
12m 44s
test-qnn-optimum-model (fp32, focalnet) / linux-job
14m 3s
test-huggingface-transformers-macos (bert|coreml_fp32_gpu|--quantize) / macos-job
9m 32s
test-qnn-optimum-model (fp32, bert) / linux-job
17m 2s
test-qnn-optimum-model (fp32, albert) / linux-job
12m 45s
test-qnn-optimum-model (fp32, roberta) / linux-job
14m 53s
test-qnn-optimum-model (fp32, distilbert) / linux-job
12m 14s
test-qnn-optimum-model (fp32, efficientnet) / linux-job
12m 43s
test-qnn-optimum-model (fp32, dit) / linux-job
13m 33s
test-qnn-optimum-model (fp32, mobilevit_v2) / linux-job
12m 46s
test-qnn-optimum-model (fp32, swin) / linux-job
14m 40s
test-qnn-optimum-model (fp32, mobilevit_v1) / linux-job
12m 59s
test-qnn-optimum-model (fp32, cvt) / linux-job
13m 17s
test-llama-runner-qnn-linux (fp32, qnn_16a16w, qnn) / linux-job
23m 58s
test-llama-runner-linux (fp32, xnnpack+custom, linux.arm64.2xlarge, executorch-ubuntu-22.04-gcc11... / linux-job
19m 46s
test-llama-runner-qnn-linux (fp32, qnn_8a8w, qnn) / linux-job
38m 25s
test-llama-runner-linux (bf16, portable, linux.arm64.2xlarge, executorch-ubuntu-22.04-gcc11-aarch64) / linux-job
13m 58s
test-llama-runner-linux (bf16, portable, linux.2xlarge, executorch-ubuntu-22.04-clang12) / linux-job
10m 51s
test-llama-runner-linux (bf16, custom, linux.arm64.2xlarge, executorch-ubuntu-22.04-gcc11-aarch64) / linux-job
14m 0s
test-llama-runner-linux (fp32, portable, linux.2xlarge, executorch-ubuntu-22.04-clang12) / linux-job
10m 42s
test-qnn-model (fp32, mv3) / linux-job
12m 25s
test-llama-runner-linux (fp32, portable, linux.arm64.2xlarge, executorch-ubuntu-22.04-gcc11-aarch64) / linux-job
13m 54s
test-llama-runner-linux (fp32, xnnpack+custom, linux.2xlarge, executorch-ubuntu-22.04-clang12) / linux-job
14m 2s
test-qnn-model (fp32, ic4) / linux-job
13m 26s
test-qnn-model (fp32, dl3) / linux-job
12m 58s
test-qnn-model (fp32, mv2) / linux-job
12m 33s
test-qnn-model (fp32, mb) / linux-job
31m 55s
test-qnn-model (fp32, ic3) / linux-job
12m 51s
test-qnn-model (fp32, conv_former) / linux-job
13m 40s
test-qnn-model (fp32, vit) / linux-job
13m 3s
test-qnn-model (fp32, w2l) / linux-job
11m 53s
test-arm-backend-vkml (test_pytest_ops_vkml) / linux-job
1h 30m 4s
test-models-windows (resnet50, portable) / windows-job
32m 8s
test-models-windows (vit, portable) / windows-job
29m 16s
test-models-windows (mv3, portable) / windows-job
30m 51s
test-models-windows (vit, xnnpack-q8) / windows-job
29m 40s
test-models-windows (mobilebert, portable) / windows-job
35m 21s
test-models-windows (mv3, xnnpack-q8) / windows-job
33m 12s
test-models-windows (emformer_transcribe, portable) / windows-job
30m 18s
test-models-windows (emformer_transcribe, xnnpack-q8) / windows-job
37m 53s
test-models-windows (mobilebert, xnnpack-q8) / windows-job
38m 26s
test-models-windows (resnet50, xnnpack-q8) / windows-job
32m 49s
test-models-macos-coreml (edsr) / macos-job
13m 23s
test-models-macos-mps / macos-job
14m 5s
test-models-macos-coreml (ic3) / macos-job
13m 14s
test-models-macos-coreml (emformer_join) / macos-job
13m 10s
test-models-macos-coreml (dl3) / macos-job
13m 27s
test-models-macos-coreml (emformer_transcribe) / macos-job
14m 9s
test-models-macos-coreml (vit) / macos-job
13m 43s
test-models-macos-coreml (efficient_sam) / macos-job
14m 29s
test-models-macos-coreml (ic4) / macos-job
13m 29s
test-models-macos-coreml (mobilebert) / macos-job
13m 53s
test-models-macos-coreml (mv3) / macos-job
13m 47s
test-models-macos-coreml (w2l) / macos-job
14m 4s
test-models-macos-coreml (mv2) / macos-job
14m 13s
test-models-macos-coreml (resnet50) / macos-job
13m 30s
unittest-release / windows / windows-job
10m 0s
test-cortex-m-e2e (mv2) / linux-job
12m 41s
unittest-release / macos / macos-job
36m 2s
unittest-release / linux / linux-job
1h 2m 54s
test-arm-ootb-linux (run_ootb_tests_tosa) / linux-job
8m 46s
test-cortex-m-e2e (mv3) / linux-job
12m 35s
test-arm-ootb-linux (run_deit_e2e_ethos_u) / linux-job
24m 7s
test-llama-runner-mac (fp32, coreml) / macos-job
16m 13s
test-arm-ootb-linux (run_ootb_tests_ethos_u) / linux-job
10m 57s
test-multimodal-macos (gemma3-4b) / macos-job
27m 53s
test-llama-runner-mac (fp32, xnnpack+custom+quantize_kv) / macos-job
16m 44s
test-llama-runner-mac (fp32, mps) / macos-job
15m 2s
test-arm-backend-ethos-u (test_smaller_stories_llama) / linux-job
23m 19s
test-arm-backend-ethos-u (test_pytest_models_ethos_u85) / linux-job
20m 31s
test-arm-backend-ethos-u (test_run_ethos_u85) / linux-job
19m 33s
test-arm-backend-ethos-u (test_run_ethos_u55) / linux-job
20m 13s
test-arm-backend-ethos-u (test_memory_allocation) / linux-job
12m 14s
test-arm-backend-ethos-u (test_pytest_models_ethos_u55) / linux-job
20m 10s
test-arm-backend-ethos-u (test_pytest_ops_ethos_u55) / linux-job
50m 16s
test-arm-backend-ethos-u (test_pytest_ops_ethos_u85) / linux-job
1h 0m 59s
test-static-llama-qnn-eval-linux
0s
2d 0h 40m 50s