]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
convert : support Glm4MoeLite (#18936)
authorddh0 <redacted>
Mon, 19 Jan 2026 22:09:20 +0000 (16:09 -0600)
committerGitHub <redacted>
Mon, 19 Jan 2026 22:09:20 +0000 (23:09 +0100)
commit1706a6d7c68fc54374b0807324079b2c6ebf674a
tree2f95653137cab0b496e005dfcc38986df0fb1729
parent959ecf7f234dc0bc0cd6829b25cb0ee1481aa78a
convert : support Glm4MoeLite (#18936)

* initial commit for branch

* add glm-4.7-flash, move tokenizer hash

* use `glm4` pretok

* silence flake8 E302 (CI)

* apply review feedback

* add <|user|> as eog

* also add EOG `<|observation|>`

* revert llama-vocab

* inherit vocab from glm4

---------

Co-authored-by: Xuan Son Nguyen <redacted>
convert_hf_to_gguf.py
convert_hf_to_gguf_update.py