]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 2e9eab8)
convert: allow using quantized Mistral weight (#17889)
authorXuan-Son Nguyen <redacted>
Wed, 10 Dec 2025 09:26:22 +0000 (10:26 +0100)
committerGitHub <redacted>
Wed, 10 Dec 2025 09:26:22 +0000 (10:26 +0100)
commit9e79b0116ebb6ff4a1ef1b42a7f2f64182ec4f10
treea4aeaf2d38ad2a1522c6a91c9812c10f5c2d396btree
parent2e9eab80c26d4b7e64f27c48b5af683c35c28742commit | diff
convert: allow using quantized Mistral weight (#17889)

* convert: allow using quantized Mistral weight

* data_torch.ndim

* update dequant fn

Co-authored-by: compilade <redacted>
---------

Co-authored-by: compilade <redacted>
convert_hf_to_gguf.py diff | blob | history
Packaging of ggml-org/llama.cpp