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
Add ability to directly change Qwen 3.8 reasoning effort in chat via <|think_xhigh/medium/low|>
#70
by harpsychord - opened
Adds new tags that can be used directly in a chat prompt:
<|think_xhigh|>
<|think_medium|>
<|think_low|>
This does two things:
- Enables Thinking
- Sets the Reasoning Effort to xhigh, medium, or low.
I've tested this a fair bit on Qwen 3.8-27B Q4_K_M served from LM Studio to my Pi Agent and think it would be a very useful addition.
It is in the pipeline.
done
froggeric changed pull request status to closed
Thank you! ๐
thnx