kavai/ qwen3.5-Gemini-Design:4b-q8_0

505 3 days ago

Qwen3.5-Gemini-Design enhances the models with an optimized system prompt for better reasoning, coding, and instruction-following performance. Smaller Models may suffer from slower speeds.

vision tools thinking 0.8b 2b 4b 9b 27b 35b 122b 
ollama run kavai/qwen3.5-Gemini-Design:4b-q8_0

Details

3 days ago

03a68846e6b7 · 5.3GB ·

model
qwen35
·
4.66B
·
Q8_0
system
You are an expert AI programming assistant, working with a user in the VS Code editor. When asked fo
license
Apache License Version 2.0, January 2004 http://www.apache.org/licenses/ TERMS AND CONDITIONS FOR US
params
{ "presence_penalty": 1.5, "temperature": 1, "top_k": 20, "top_p": 0.95 }

Readme

Qwen3.5-Gemini-Design

Summary: Qwen3.5-Gemini-Design is a prompt-optimized variant of the Qwen3.5 family focused on structured design workflows, layout reasoning, and high-quality visual and UI generation prompts.

Overview

Qwen3.5-Gemini-Design builds on the strong foundation of the Qwen3.5 models with a refined system prompt tailored for design-oriented tasks. The goal is to improve performance in UI/UX generation, layout planning, visual prompt engineering, and structured creative workflows while maintaining the reasoning strength of the base model.

Highlights

  • Optimized for UI/UX and product design workflows
  • Stronger structured output for layout and component descriptions
  • Improved prompt generation for Gemini-style design pipelines
  • Better reasoning for visual hierarchy and spacing logic
  • Lightweight system-layer modification (no weight changes)

Use Cases

  • UI/UX wireframe generation
  • Figma and layout prompt drafting
  • Design system creation
  • Component hierarchy planning
  • Landing page structure generation
  • Visual prompt engineering for Gemini models

Usage

ollama run kavai/qwen3.5-Gemini-Design

Design Philosophy

This release focuses on improving design-centric reasoning through prompt engineering rather than modifying model weights. By injecting a structured design-focused system layer, Qwen3.5-Gemini-Design enhances layout understanding, component grouping, and visual instruction clarity.

Benchmarks

Formal design-task benchmarks coming soon. Model speeds may be slower than expected due to the system prompt.

Notes

  • No architectural changes to base model
  • Fully compatible with Qwen3.5 tooling
  • Optimized for design and creative workflows
  • Works well with Gemini-style prompt pipelines
  • Recommended to use on GPU

License

Follows the original Qwen3.5 model license.