unsloth

/

Qwen3-30B-A3B-Instruct-2507-GGUF

like 266

Follow

Unsloth AI 11.1k

I found that you named the architecture qwen3moe while the original one is qwen3_moe.
It makes it incompatible with several frameworks, including Transformers.

Would it be possible to rename it with the standard name? (or maybe this is not possible and something hardcoded in llama.cpp?)

same issue

Yes, it is not possible to load using vllm 10.0.1.1

Value error, The checkpoint you are trying to load has model type qwen3moe but Transformers does not recognize this architecture. This could be because of an issue with the checkpoint, or because your version of Transformers is out of date.

It there a way to change gguf file to rename it?

Or maybe somehow change a vllm code to duplicate config for Qwen3MoeForCausalLM under qwen3moe ?