Instructions to use froggeric/Qwen-Fixed-Chat-Templates with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use froggeric/Qwen-Fixed-Chat-Templates with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir Qwen-Fixed-Chat-Templates froggeric/Qwen-Fixed-Chat-Templates
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Usage on Ornith-1.5-35B-A3B
Can this template be used on the Ornith-1.5-35B-A3B finetune also?
Ornith is the same family of model, and this template is working well for me on https://huggingface.co/dealignai/Ornith-1.5-35B-A3B-CRACK-GGUF - I also use it on https://huggingface.co/outsourc-e/Qwen3.8-27B-Unleashed-GGUF
Both have been great so far with this template for agentic and chat with reasoning on, and generating game json outputs with reasoning off.
I have an issue with the "auto_disable_thinking_with_tools: true" option when used with Ornith-1.5-35B-A3B and Hermes.
The tool calls are sent in xml eventhough I set the following in hermes config:
extra_body:
chat_template_kwargs:
tool_call_format: json
auto_disable_thinking_with_tools: true
Ornith 1.5 9B and Qwen 3.8 27B work with these chat template kwargs