Hugging Face's logo

froggeric
/
Qwen-Fixed-Chat-Templates

MLX
jinja
chat-template
qwen
qwen3.5
qwen3.6
lm-studio
llama.cpp
tool-calling
thinking
Model card Files
xet
 Community
4
froggeric commited on
Commit
58eb7af
·
verified ·
1 Parent(s): 2179960

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -15,7 +15,7 @@ tags:
15
 
16
  # Fixed Chat Templates for Qwen 3.5 & 3.6
17
 
18
- > **2026-05-05** — Reviewed against community merged templates (allanchan339, fakezeta). Confirmed all useful features already present; `from_json` string-arg parsing not portable to C++ engines. Documented auto-close unclosed thinking.
19
 
20
  Drop-in Jinja templates that fix rendering errors, token waste, and missing features in the official Qwen chat templates. Works in LM Studio, llama.cpp, vLLM, MLX, oMLX, and any engine that supports HuggingFace Jinja templates.
21
 
 
15
 
16
  # Fixed Chat Templates for Qwen 3.5 & 3.6
17
 
18
+ > **2026-05-05** — Reviewed against community merged templates (allanchan339, fakezeta). Confirmed all useful features already present; `from_json` string-arg parsing not portable to C++ engines. Added auto-close unclosed thinking, thanks to allanchan339.
19
 
20
  Drop-in Jinja templates that fix rendering errors, token waste, and missing features in the official Qwen chat templates. Works in LM Studio, llama.cpp, vLLM, MLX, oMLX, and any engine that supports HuggingFace Jinja templates.
21