Align with OFPBadWord pattern: FastAPI app, clean routes, no ZeroGPU

Optimize Qwen3 for fast social chat: disable thinking mode, cap tokens

er gpu changes

added smaller qwne model

decoration try for zero GPU

fix

fix

GPU enabled

qwen 3 through transformers

UPdate your project to use Qwen3-4B-Instruct-2507 and remove all other model references

initial commit