Fix FP8 fallback in cuBlas - SemanticDiff

Commit

6 days ago

Fix FP8 fallback in cuBlas

References

#28413 - CUDA: Add OCP FP8 support

#28898 - ggml: support quant scales for fp8 and nvfp4

#29791 - vulkan: add fp8 and scaled matmul support

Author

ORippler

ORippler

Committer

ORippler

ORippler

Parents

Loading