Skip to main content
Endpoint: POST https://fal.run/bria/fibo-lite/generate Endpoint ID: bria/fibo-lite/generate

Try it in the Playground

Run this model interactively with your own prompts.

​
Quick Start

import fal_client

def on_queue_update(update):
    if isinstance(update, fal_client.InProgress):
        for log in update.logs:
           print(log["message"])

result = fal_client.subscribe(
    "bria/fibo-lite/generate",
    arguments={},
    with_logs=True,
    on_queue_update=on_queue_update,
)
print(result)

​
Examples

several coffee beans floating independently in mid air, the subject is red, in a red photo, red and black duotone image, red wash, red light, red lens, extreme close up, a dark black empty background, closeup, dramatic lighting with a strong red hue that gives the image a moody and intense atmospher…
Generated image: several coffee beans floating independently in mid air, the subject is red, in a
Example output from Fibo Lite
Example output from Fibo Lite

​
Input Schema

​
prompt
string
The prompt to generate.
​
structured_prompt
StructuredPrompt
The structured prompt to generate.
​
image_url
string
Input image URL
​
seed
integer
default:"7"
Seed for the random number generator. Default value: 7
​
aspect_ratio
AspectRatioEnum
default:"1:1"
Aspect ratio. Options: 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9 Default value: "1:1"Possible values: 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9
​
negative_prompt
string
default:""
Negative prompt for image generation. Default value: ""
​
steps_num
integer
default:"8"
Number of inference steps. Default value: 8Range: 4 to 30
​
sync_mode
boolean
default:"false"
If true, returns the image directly in the response (increases latency).

​
Output Schema

​
image
Image
required
Generated image.
​
images
list<object>
Generated images.
​
structured_prompt
StructuredPrompt
required
Current prompt.

​
Input Example

{
  "prompt": "Oil painting of a fluffy, wide-eyed cat sitting upright, holding a small wooden sign reading \"Feed Me.\" Rich textures, dramatic brushstrokes, warm tones, and vintage charm.",
  "seed": 7,
  "aspect_ratio": "1:1",
  "negative_prompt": "",
  "steps_num": 8,
  "sync_mode": false
}

​
Output Example

{
  "image": {
    "url": "",
    "content_type": "image/png",
    "file_name": "z9RV14K95DvU.png",
    "file_size": 4404019,
    "width": 1024,
    "height": 1024
  },
  "structured_prompt": {
    "short_description": "",
    "objects": [
      {
        "description": "",
        "location": "",
        "relationship": ""
      }
    ],
    "background_setting": "",
    "lighting": {
      "conditions": "",
      "direction": ""
    },
    "aesthetics": {
      "composition": "",
      "color_scheme": "",
      "mood_atmosphere": "",
      "aesthetic_score": "",
      "preference_score": ""
    },
    "text_render": [
      {
        "text": "",
        "location": "",
        "size": "",
        "color": "",
        "font": ""
      }
    ],
    "context": "",
    "artistic_style": ""
  }
}
Overview Fibo Lite, the new addition to the Fibo model family, allows generating high-quality images with the same controllability of the JSON structured prompt, while using significantly fewer inference steps, resulting in significantly improved latency. This is a two-stage distilled model for the FIBO text-to-image model, combining:
  1. CFG Distillation: First, we distill classifier-free guidance into the model, enabling inference with Guidance Scale = 1.0 (skipping the negative prompt pass).
  2. SCFM (Shortcutting Flow Matching): On top of the CFG-distilled merged model, we apply velocity field self-distillation to enable efficient few-step sampling.

​
🔑 Key Benefits

  • Two-Stage Distillation: Combines CFG distillation with SCFM for maximum efficiency—the SCFM was trained on top of the already CFG-distilled merged model.
  • Few-Step Generation: SCFM enables efficient sampling in significantly fewer inference steps.
  • No CFG Overhead: Running at guidance_scale=1 means calculating the noise prediction only once per step instead of twice.
  • Quality Tradeoff: As a distillation approach, there is a slight quality degradation compared to the full model at CFG=5, but the speed gains make it ideal for rapid iteration and production workflows.
  • Drop-in Replacement: Works seamlessly with existing FIBO workflows—just set guidance_scale=1.0.
  • Memory Efficient: Minimal additional GPU memory overhead.

​
Limitations

  • steps_num range: 4 to 30