]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 614d3b9)
convert-hf : support direct Q8_0 conversion (#7234)
authorcompilade <redacted>
Mon, 13 May 2024 18:10:51 +0000 (14:10 -0400)
committerGitHub <redacted>
Mon, 13 May 2024 18:10:51 +0000 (14:10 -0400)
commitee52225067622babc277371511b8124884e1c797
tree8150564487416fc038952c8b85f1462f3b1c98cftree
parent614d3b914e1c3e02596f869649eb4f1d3b68614dcommit | diff
convert-hf : support direct Q8_0 conversion (#7234)

* convert-hf : support q8_0 conversion

* convert-hf : add missing ftype

This was messing with the checksums otherwise.

* convert-hf : add missing ftype to Baichuan and Xverse

I didn't notice these on my first pass.
convert-hf-to-gguf.py diff | blob | history
gguf-py/gguf/__init__.py diff | blob | history
gguf-py/gguf/gguf_writer.py diff | blob | history
gguf-py/gguf/lazy.py diff | blob | history
gguf-py/gguf/quants.py [new file with mode: 0644] blob
Packaging of ggml-org/llama.cpp