Skip to content

dflash: fix zero acceptance rate on CUDA (F16 tensor read + encoder overflow)#2

Open
fuzzballb wants to merge 1 commit into
ruixiang63:dflashfrom
fuzzballb:pr-22105
Open

dflash: fix zero acceptance rate on CUDA (F16 tensor read + encoder overflow)#2
fuzzballb wants to merge 1 commit into
ruixiang63:dflashfrom
fuzzballb:pr-22105

Enhance your code review process with GitHub Actions

GitHub Actions make it easy to automate all your software workflows, now with world-class CI/CD.
Build, test, and deploy your code right from GitHub. Learn more about GitHub Actions.

Linux, macOS, Windows, and containers
Linux, macOS, Windows, and containers
Matrix builds
Matrix builds
Any language
Any language
Live logs
Live logs
Built-in secret store
Built-in secret store
Multi-container testing
Multi-container testing