Gradio shrinks Qwen-Image 2.1's 9B prompt rewriter to 0.8B that runs on laptops
Gradio · x · 2026-09-23
The Gradio team found that Qwen-Image 2.1 draws its best images only after a 9B "prompt rewriter" expands your request: 20 GB in bf16, a 1,700-word system prompt, and 1,600 thinking tokens per answer. They compressed it to 0.8B so it fits on a laptop, publishing the details in a thread.
Related event: Gradio Distills Qwen's 9B Prompt Rewriter to 0.8B, Fully Open-Sources It(4 posts)→
More from Multimodal
- Izwi: open-source offline voice cloning in Rust, runs on 16GB M1 Pro — zinyando · 2026-11-12
- MiniMax H3 Image generates 3D game scenes in 10s, over 3x faster than GPT Image — HeyToha · 2026-11-12
- Dev releases local world model turning any image into interactive video, runs on RTX 3060 — lucidml_lover · 2026-11-12
- Ex-Meta CV researcher builds AdaL v2, a chat-driven video agent harness for GTM — Scobleizer · 2026-11-12
- MiniMax H3 image gen runs locally on AMD RX 9070, 25-30s per 4MP image — omglionheaded · 2026-11-12
- One reference, one prompt, three image models: side-by-side Nano Banana Pro vs GPT Image 2.6 test — gizakdag · 2026-11-12