Eval bug: Transformers does not recognize this architecture

[Xiao-hb]

Name and Version

Latest version

Operating systems

Windows

GGML backends

CPU

Hardware

CPU

Models

No response

Problem description & steps to reproduce

Based on HuggingFace's pretrained SigLiP2 and Qwen3, I locally trained a combination of the two to build my own VLM. The config.json file shows "architecture": "VLM".

When using convert_hf_to_gguf.py for conversion, it prompts: "ERROR: hf-to-gguf: Model VLM is not supported"

How can I solve this? Thank you very much.

First Bad Commit

No response

Relevant log output

Logs