Israel8B parametersText to imageJSON-native

Explore FIBO

Create and iteratively refine controllable images with BRIA's JSON-native text-to-image model trained entirely on licensed data.

Model

briaai/FIBO

Input

Configure your request

Generation mode

Prompt or refinement instruction

Structured JSON prompt Optional

Optional FIBO JSON-native visual description for precise control.

Inspiration image Optional

Used in Inspire mode to derive a structured visual prompt.

Negative prompt Optional

Describe visual properties to avoid.

Parameters

Width

Height

Inference steps

Guidance scale

Number of images

Seed

Custom timesteps

Optional custom diffusion schedule.

Maximum prompt tokens

FIBO supports structured prompts up to 3,000 tokens.

Output type

Patch-based generationUse the pipeline’s patching mode for generation.

Joint-attention options (JSON)

Optional keyword arguments passed to the joint-attention processor.