Add CUDA kernel support for 4-bit quantization with blocksize=32 #1554
matthewdouglasbuild_pr_documentation.yml
on: pull_request
build
/
build_pr_documentation
2m 51s
Artifacts
Produced during runtime
| Name | Size | Digest | |
|---|---|---|---|
|
doc-build-artifact
|
1.17 MB |
sha256:15603159c395de3e58101a62ca52800b634957dccbf1f3471ffd51e046bff413
|
|