Qwen3-4B-Instruct-2507 error converting/quantizing

[0wwafa]

Qwen/Qwen3-4B-Instruct-2507

python /content/llama.cpp/convert_hf_to_gguf.py --outtype f16 Qwen3-4B-Instruct-2507 --outfile Qwen3-4B-Instruct-2507.f16.gguf

Traceback (most recent call last):
File "/content/llama.cpp/convert_hf_to_gguf.py", line 32, in
from mistral_common.tokens.tokenizers.base import TokenizerVersion
ModuleNotFoundError: No module named 'mistral_common'

version: llama.cpp binaries version b6139

[CISC]

Requirements changed, you have to update your venv from requirements/requirements-convert_hf_to_gguf.txt.

[0wwafa]

my bad. I noticed immediately after.. sorry.

[arch-btw]

@CISC I'm not trying to cause more issues than what's going on here, but I noticed that I also need mistral-commons to convert Hunyuan-4B-Instruct now, is this by design?

I'm just trying to understand why we need a mistral package for non-mistral models.

Thank you for your time.

[CISC]

@arch-btw It was simpler this way. Conditionally importing it got messy and it's not a big dependency.

[arch-btw]

@CISC Got it, thank you.