]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
convert : support latest mistral-common (fix conversion with --mistral-format) (...
authorSmartestWashingMachine <redacted>
Wed, 3 Dec 2025 20:15:04 +0000 (07:15 +1100)
committerGitHub <redacted>
Wed, 3 Dec 2025 20:15:04 +0000 (21:15 +0100)
commit424c5794557597c8bbd9ea318570962b9ad00e22
treeb65fc5f857041e5816cd4099cf00ac7c45ce2f7b
parente9f9483464e6f01d843d7f0293bd9c7bc6b2221c
convert : support latest mistral-common (fix conversion with --mistral-format) (#17712)

* fix convert_hf_to_gguf.py failing with --mistral-format using later mistral-common versions.

* use get_one_valid_tokenizer_file from mistral-common if available and fallback to old logic otherwise.

* use file name instead of file path for get_one_valid_tokenizer_file.

* fix --mistral-format tokenizer file failing for tokenizers in subdirectories.

* move get_one_valid_tokenizer_file import to avoid nested try-except.
gguf-py/gguf/vocab.py