Skip to content
8 GPU Integration Test on H100

Add blockwise FP8 training recipe for linears and MoE grouped experts #10908

Sign in to view logs

Add blockwise FP8 training recipe for linears and MoE grouped experts

Add blockwise FP8 training recipe for linears and MoE grouped experts #10908

Triggered via pull request July 9, 2026 17:47
@iamzainhudaiamzainhuda
synchronize #3891
iamzainhuda:fp8-blockwise-linears-and-moe
Status Skipped
Total duration 1s
Artifacts –

integration_test_8gpu_h100.yaml

on: pull_request
set-matrix  /  set
set-matrix / set
set-matrix-rocm  /  set
set-matrix-rocm / set
Matrix: build-test
Waiting for pending jobs
Matrix: build-test-rocm
Waiting for pending jobs
Fit to window
Zoom out
Zoom in