symmetric quantization to FBGEMM prefill token-wise FP8 (fixed) #5529
ColinPepplerfbgemm_gpu_benchmark_cpu.yml
on: pull_request
Matrix: build_artifact
Matrix: benchmark_artifact
Artifacts
Produced during runtime
| Name | Size | Digest | |
|---|---|---|---|
|
fbgemm_gpu_nightly_cpu_arm_gcc_py3.13.whl
|
4.22 MB |
sha256:7dd72276343b9db2adab528aa95091d3f58e94e67f2c623e93fa5fc57b797c0c
|
|
|
fbgemm_gpu_nightly_cpu_x86_gcc_py3.13.whl
|
5.36 MB |
sha256:6b2440bf5e6ceb804f7d060efdc5eb11a3920e40d3333c2bba2209c7251e3406
|
|
|
fbgemm_gpu_traces_arm_gcc_py3.13_cpu.zip
|
238 KB |
sha256:c0bcbd0c0e7c18ee8d603f28d060ffc122843cf0123f5265a6a4e9b56b851f34
|
|
|
fbgemm_gpu_traces_x86_gcc_py3.13_cpu.zip
|
238 KB |
sha256:788a90e6b4a85f58a82cf508ebccbef6aa528aae1d7646171fef7f702d670fb1
|
|