huihui-ai

/

Huihui-Qwen3-VL-8B-Instruct-abliterated

like 131

I know the gguf was broken at least for ollama. But I wanted to ask if you can release a new gguf file that does work on ollama this time? Or is this not planed. gguf models would be very usefull for the 32b and 30ba3b models. in ollama you can import this models via savetensors but this models require way to much storage in safetensors format. gguf in q8_0 would be realy usefull.

The Qwen3-VL version should have different GGUF (Generative Pre-trained Transformer) models for llama.cpp and
ollama. You can obtain the GGUF for ollama from the saved file directory where you pulled the ollama source code.
We haven’t tested it yet, so we don’t know if it will meet your requirements.