🤗 Open Model Playground

Text model: Qwen/Qwen3-4B-Instruct-2507
Image model: diffusers/FLUX.2-dev-bnb-4bit
Device: cuda

Text Generation with Qwen3 + Transformers

System message

User prompt

max_new_tokens

32 1024

temperature

0 1.5

top_p

0.1 1

Generated answer

Metadata

Runtime profile

Colab A100: Qwen3 + SDXL
HF Space ZeroGPU: Qwen3 + FLUX.2-dev-bnb-4bit