✦
Summary
Fibo 1.5 is Bria JSON-native image model for controlled visual generation. It describes composition through structured attributes such as lighting, camera, color, and scene content, giving creative workflows a more explicit way to revise an image without letting every unrelated detail drift.
Key Features
Structured JSON explicitly represents visual attributes such as camera, lighting, and composition.
A vision-language model can expand a short idea into a detailed structured description.
The refinement workflow targets selected attributes while preserving the broader scene description.
An image can be converted into a structured prompt for related outputs.
Fibo 1.5 recommends four to six steps without classifier-free guidance.
The guidance-free sampling setup does not use negative prompts.
The authors recommend the base FIBO checkpoint for fine-tuning instead.
Noncommercial weight access and commercial licensing have separate terms.
About Fibo 1.5
The eight-billion-parameter FIBO family is trained on licensed data and detailed structured captions. Version 1.5 applies DMD followed by DMD-R refinement to generate in four to six steps without classifier-free guidance. A vision-language model can expand short instructions or image references into the structured prompt.
Fibo 1.5 supports generation, targeted refinement, and image-inspired creation through local inference, ComfyUI, and hosted integrations. Weights are gated for noncommercial access, while commercial use requires separate licensing. The distilled checkpoint is not intended for fine-tuning; users needing adaptation should start from the base model.