]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: d161ea7)
convert : add compressed-tensors NVFP4 support (#21095)
authorMichael Wand <redacted>
Mon, 25 May 2026 12:16:11 +0000 (08:16 -0400)
committerGitHub <redacted>
Mon, 25 May 2026 12:16:11 +0000 (14:16 +0200)
commita4d2d4ae41cf6a97d9e828e7240c4045e9b2e2b1
tree58ce8208c4cd050667cc395f2e31754af3b63285tree
parentd161ea707192f9510288561104a8caf8e1829edfcommit | diff
convert : add compressed-tensors NVFP4 support  (#21095)

* Refactored Compressed Tensors NVFP4 support for new base.py

* Support compressed-tensors NVFP4 conversion

* Moved Qwen MTP remap into filter_tensors

* simplify

* pathlib no longer used

---------

Co-authored-by: Sigbjørn Skjæret <redacted>
conversion/base.py diff | blob | history
conversion/qwen.py diff | blob | history
Packaging of ggml-org/llama.cpp