MLX
jinja
chat-template
qwen
qwen3.5
qwen3.6
qwen3.8
llama.cpp
lm-studio
vllm
tool-calling
thinking
token-efficient
Instructions to use peculiar-ragdoll/Qwen-Sharp-Chat-Templates with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use peculiar-ragdoll/Qwen-Sharp-Chat-Templates with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir Qwen-Sharp-Chat-Templates peculiar-ragdoll/Qwen-Sharp-Chat-Templates
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Froggeric just release the new Qwen-Fixed-Chat-Templates v22
#2
by lawlietr - opened
Hi, Your templates real amazing! Help my code agent alot.
Hope the Qwen-Sharp-Chat-Templates get that update too.
Thank you.
Thank you for your kind words, and for the heads up! 🙏 New version of my template is up on the repo now :) I made some edits to the Froggeric one, turning off the "xhigh" default effort level, making thinking effort instructions only activate when passed explicitly, so current behavior stays consistent! Tested it on oMLX and llama.cpp. Enjoy!
@peculiar-ragdoll
Https://huggingface.co/Moore2877/Qwen-Fixed-Chat-Templates-llamacpp
Is there something in this release that could improve yours?