llama.cpp
7cbec596 - Fix conversion edge-case for FP8 block-scales

Commit

1 day ago

Fix conversion edge-case for FP8 block-scales

References

#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