Text Generation
Safetensors
English
qwen3_5
code
agentic
distillation
demonstration
evaluation
conversational
Instructions to use DJLougen/Qwable-5-27B-Coder with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Inference
Does not perform well
#1
by Hypersniper - opened
Does not perform as well as the other Qwable 27b release in html/javascript with image and instruction following. Requires multi follow up conversations to fix the initial code. Performs a little worse than the original release of Qwen. It might of lost some precision to in html and JavaScript coding during the fine-tune. Perhaps it’s better at python. Thanks for releasing the model though great job!
I had the same experience. Model card explains why it was crap. 😂
i crap max