Introducing FIBO: Structured Control for Text-to-Image Generation

Team fal

• 2 min read
Introducing FIBO: Structured Control for Text-to-Image Generation

We're excited to launch Bria's latest text-to-image model, FIBO, on fal!

FIBO turns simple text prompts into structured JSON schemas, giving creators predictable, controllable visuals at scale. Trained on richly detailed captions (often 1,000+ words), it deeply understands lighting, composition, color, camera settings, and depth of field.

Why FIBO Matters

Most image models interpret open-ended text prompts, leaving room for variation. FIBO instead converts prompts into a structured JSON schema, where outputs are consistent, reproducible, and controllable.

As the first JSON-native image generation model, FIBO brings traceability, auditability, and precise editability to every render — purpose-built for agentic, automated, and professional design workflows.

Key Capabilities

  • Structured JSON prompting: Describe scenes with precision—lighting, composition, DoF, camera specs—all within a schema up to 1,000 words.
  • Iterative controlled generation: Start from a short prompt or refine from a detailed one. Each edit updates only what you change.
  • Disentangled control: Adjust one variable (like “backlight” or “85mm”) without distorting the rest.
  • LLM-guided creativity: FIBO expands short prompts into rich JSONs, offering both inspiration and control.
  • Rights-clear, production-ready: Built entirely on licensed data with full enterprise indemnity.

What can you do with FIBO?

Generate

Start with a quick idea. FIBO’s language model expands your short prompt into a rich, structured JSON prompt, then generates the image. You get both the image and the expanded prompt.

Refine

Continue from a detailed structured prompt and add a small instruction - for example, “backlit,” “85 mm,” or “warmer skin tones.” FIBO updates only the requested attributes, re-generates the image, and returns the refined prompt alongside it.

Inspire

Provide an image instead of text. FIBO’s vision-language model extracts a detailed prompt describing it, combines that with your creative intent, and generates new, related images - ideal for exploring new directions without altering the original.

Get Started with FIBO

You can try FIBO now in the fal Playground or integrate it via API for both text-to-image and image-to-image generation.

Stay tuned to our Reddit, blog, Twitter, or Discord for the latest updates on generative media and the new model releases!