$

Cost per second

For enterprise pricing and custom weights or models

Bria Vector Graphics Model

What is Bria Vision?

Bria Vision is a next-generation generative AI model designed for creating both raster and scalable vector graphics from text prompts or reference images. Built for developers, designers, and product managers, this unified pipeline supports:

  • Text-to-Image (base, fast, HD) for photo-realistic or illustrative assets
  • Text-to-Vector (base, fast) for clean, editable SVG-style icons, logos, and illustrations

With advanced guidance methods like ControlNets and Image Prompt Adapters, Bria Vision offers fine-grained control over style, resolution, aspect ratio, and generation speed—ideal for commercial-grade asset production at scale.

Key Features

  • Dual Pipelines: Generate high-quality raster images (text-to-image) or crisp vector outputs (text-to-vector)
  • Mode Selection:
    • base – Balanced quality & speed
    • fast – Quick iterations and mockups
    • hd – Ultra-detailed, production-ready visuals
  • Advanced Conditioning: Support for ControlNets (edges, depth) and Image Prompt Adapters (style transfer, recoloring)
  • Custom Resolutions & Aspect Ratios: Choose from presets like 1:1, 16:9, 4:3, or custom values for social media, presentations, and UI assets
  • Deterministic Seeds: Reproduce or slightly vary outputs using seed values (1–999999)
  • Negative Prompting: Exclude unwanted elements (e.g., “no dark colors”) for tighter composition control

Best Use Cases

  • Logo & Icon Design: Generate sharp vector graphics for branding and UI kits
  • Marketing & Social Media: Create HD visuals in 16:9 or custom aspect ratios for ads and posts
  • Illustrations & Concept Art: Produce stylized art or photo-realistic scenes with advanced text guidance
  • Prototyping & Mockups: Rapidly iterate UI components, stickers, and banners using the fast mode
  • Scalable Asset Libraries: Build large collections of consistent imagery for websites, apps, and products

Prompt Tips and Output Quality

  1. Craft Detailed Prompts: Describe composition, lighting, and style. E.g., “A minimalist vector icon of a fox, flat colors, 1:1 ratio.”
  2. Choose the Right Mode: Use fast for drafts, base for balanced renders, and hd for final assets.
  3. Aspect Ratio: Select 1:1 for icons, 16:9 for banners, or customize via the dropdown.
  4. Seed Control: Set a seed to reproduce results; tweak by ±1 for variations.
  5. Negative Prompts: Filter out unwanted elements—e.g., “no text overlays.”
  6. Guidance Scale & Steps: Increase text_guidance_scale (1–10) for strict adherence; set steps_num (20–50) for smoother details.
  7. Enhancements & Moderation: Toggle Enhance Image for clarity; keep Prompt Content Moderation enabled for compliance.

FAQs

Q: How do I generate editable vector graphics?
A: Select the text-to-vector pipeline and choose base or fast mode. Use detailed prompts like “minimalist icon” and pick 1:1 aspect ratio.

Q: Which mode yields the highest detail?
A: Use hd mode. It performs more diffusion steps and texture enhancements for production-quality outputs.

Q: Can I reuse the same image output?
A: Yes. Set the seed parameter (1–999999) to the same value to replicate results exactly.

Q: How do I control composition?
A: Combine aspect_ratio presets with negative_prompt fields (e.g., “no background clutter”) and adjust guidance_scale.

Q: Are there content moderation safeguards?
A: Bria Vision includes Prompt Content Moderation and Content Moderation checkboxes to ensure compliance with standard guidelines.