Skip to content
Back to pull request #7927

[Excutorch][Llama] Decouple input sequence length from kv cache context length #7556

Sign in to view logs

[Excutorch][Llama] Decouple input sequence length from kv cache context length

[Excutorch][Llama] Decouple input sequence length from kv cache context length #7556

Run time

Learn about OS pricing on GitHub Actions
Job Run time
gather-models
6s
test-arm-backend-delegation / linux-job
20m 40s
test-arm-reference-delegation / linux-job
26m 43s
test-coreml-delegate / macos-job
51m 8s
test-custom-ops-macos (cmake) / macos-job
10m 5s
test-demo-backend-delegation (buck2) / linux-job
7m 3s
test-apple-model / macos-job
15m 23s
test-demo-backend-delegation (cmake) / linux-job
6m 19s
test-selective-build-macos (cmake) / macos-job
10m 17s
test-huggingface-transformers (google/gemma-2-2b) / linux-job
48m 20s
test-llama-runner-mac (bf16, custom) / macos-job
13m 46s
test-llama-runner-mac (bf16, portable) / macos-job
11m 57s
test-llama-runner-mac (fp32, coreml) / macos-job
12m 56s
test-llama-runner-mac (fp32, mps) / macos-job
13m 3s
test-llama-runner-mac (fp32, portable) / macos-job
12m 3s
test-llama-runner-mac (fp32, xnnpack+kv+custom) / macos-job
12m 36s
test-llama-runner-mac (fp32, xnnpack+custom+quantize_kv) / macos-job
12m 54s
test-llama-runner-qnn-linux (fp32, qnn_16a16w, qnn) / linux-job
47m 20s
test-llama-runner-qnn-linux (fp32, qnn_8a8w, qnn) / linux-job
32m 17s
test-pybind-build-macos (cmake) / macos-job
9m 23s
test-qnn-model (fp32, dl3) / linux-job
9m 57s
test-qnn-model (fp32, ic3) / linux-job
9m 39s
test-qnn-model (fp32, ic4) / linux-job
9m 33s
test-qnn-model (fp32, mv2) / linux-job
8m 40s
test-qnn-model (fp32, mv3) / linux-job
8m 28s
test-qnn-model (fp32, vit) / linux-job
9m 13s
test-models-macos (cmake, mv3, portable, macos-m1-stable, 90) / macos-job
13m 10s
test-models-macos (cmake, mv3, xnnpack-quantization-delegation, macos-m1-stable, 90) / macos-job
10m 31s
test-models-macos (cmake, vit, portable, macos-m1-stable, 90) / macos-job
11m 14s
test-models-macos (cmake, vit, xnnpack-quantization-delegation, macos-m1-stable, 90) / macos-job
11m 3s
7h 55m 47s