Bria FIBO 2.0 Uncensored

Also known as: FIBO 2.0 · FIBO 2

Bria FIBO 2.0 is the newest generation of FIBO, the image model family from Bria AI. Bria built FIBO around structured prompting: the model learned from long, detailed captions that describe lighting, composition, color and camera one attribute at a time, and Bria presents it as a model for controlled, repeatable pictures. Bria also says its training data is licensed. This page runs FIBO 2.0 for both text-to-image and editing.

Text-to-image gives you 1K or 2K output, with 2K as the default, and both sizes cost the same. 2K covers the standard square, wide and tall shapes; 1K adds a set of extra-wide and extra-tall frames for long banners and narrow vertical layouts. Since the price never moves with size or shape, the decision is only about where the picture will be shown.

Bria's own tools can take a full JSON description of a scene. Here you write a plain-text prompt, and it reaches the model exactly as typed, with nothing rewriting it on the way. You keep most of the benefit by writing the way FIBO was trained: name the subject, then wardrobe, pose, setting, light, lens and palette, each as a short plain statement. A seed lets you repeat a result and then change a single attribute.

Editing takes one to four reference pictures, and extra references are not charged. The edited image keeps the proportions of the first picture, so trim that one to the shape you need before uploading. For instruction edits that hold a face closely, compare Qwen Image Edit 2511; for merging several photos at high resolution, look at Wan 2.7 Image. The previous generation, Bria FIBO 1.5, remains available at the same price.

On Vellria, Bria FIBO 2.0 is uncensored for fictional adult content: nudity and explicit scenes between adults are allowed, and nothing rewrites your prompt. An automatic check refuses prompts that sexualize minors, and real people may not be shown without their consent.

Bria FIBO 2.0 outputs with the exact prompts behind them.

• 

• 

• 

• 

• 

• 

• 

• Before

  After

See also: uncensored AI image generator · uncensored image editor.

Bria FIBO 2.0 is the current generation of Bria's FIBO line on Vellria. Its strengths are prompts written as plain lists of attributes, a choice of output size at one price, and edits that take several references without any extra charge.

Bria FIBO 2.0 costs 10 credits over the API, the same rate as in Studio. Model id vellria-bria-fibo-2-0 on POST /v1/generate/image.

curl https://vellria.com/v1/generate/image \
  -H "Authorization: Bearer vll_..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "vellria-bria-fibo-2-0",
    "prompt": "…",
    "size": "2048*2048"
  }'

Parameters, polling and uploads: API documentation · NSFW image API.

Turn a still into a clip with one of our video models.

Pruna P-Video 2 Pro Uncensored

For the take you keep: 5–15 s clips with sound at 480p or 768p, or start from a still.

Video

from 12 credits / s

Pruna P-Video 2 Uncensored

From a one-second beat to a 20-second take, with sound, at 720p or 1080p; animates stills too.

Video

from 8 credits / s

LTX 2.5 Pro Uncensored

Top-quality takes with sound, 6, 8 or 10 s, from 720p up to 4K, landscape or portrait; animates stills too.

Video

from 35 credits / s