Explore BRIA 2.3

Provides the exact transcript the speech model will read aloud. Write the words, punctuation, and language you want spoken; punctuation can influence pauses and phrasing, while pronunciation still depends on the model’s training data.

Describes unwanted subjects, styles, defects, or artifacts such as blur or watermarks. It steers generation away from those concepts but is not a guaranteed content filter.

Provides an optional prompt to the pipeline’s second text encoder. Use it to give that encoder different emphasis or detail; leaving it empty normally reuses the main prompt so both encoders receive the same instruction.

Provides avoidance instructions to the second text encoder. Leave it empty to reuse the main negative prompt, or set it when the second encoder should suppress different concepts or visual qualities.

Sets the vertical resolution of generated images or video frames in pixels. Larger values create more detail but require more memory and processing time; keep the height compatible with the model’s supported dimensions and desired aspect ratio.

Sets the horizontal resolution of generated images or video frames in pixels. Larger values create more detail but require more memory and processing time; combine it with height to choose the aspect ratio.

Controls how many iterative denoising passes produce the final media. More steps can refine detail and consistency but increase generation time, with diminishing quality gains beyond the model’s useful range.

Overrides the scheduler’s normal diffusion timestep sequence with a custom descending list. This is an advanced reproducibility and scheduler-tuning control; use a schedule supported by the selected scheduler and do not combine it with custom sigmas.

Overrides the scheduler’s default noise-level sequence with custom sigma values. This changes how denoising effort is distributed and can strongly affect quality; use values designed for the selected scheduler and do not combine them with custom timesteps.

Sets the strength of classifier-free guidance toward the text prompt. Higher values usually improve literal prompt adherence but can reduce natural motion or create artifacts; lower values give the model more visual freedom.

Rescales classifier-free guidance to reduce overexposure, excessive contrast, or color saturation at higher guidance strengths. Zero disables rescaling, while larger values apply more correction and may soften prompt adherence.

Requests multiple independent image variations from the same prompt in one run. Each additional image increases compute and memory use; random generator state or seeds determine how much the results differ.

Controls the amount of additional stochastic noise used by DDIM-compatible schedulers. Zero is deterministic for a fixed seed, while higher values can create more variation; schedulers that do not accept eta ignore this setting.

Uses an earlier CLIP text-encoder layer instead of the final layer when interpreting the prompt. Some model families were trained with a particular skip value, but arbitrary changes can weaken prompt understanding; leave it unset unless the model documentation recommends one.

Initializes the pipeline’s random-number generator. Reusing the same seed, prompt, model, and settings usually reproduces the same output, while changing it creates a new variation.

Chooses the structure of the pipeline result. Enabled returns an object with named properties that is easier to read; disabled returns a positional tuple that may suit older or lower-level integrations.

Chooses the data representation returned by the pipeline. Image frames are convenient for viewing, NumPy arrays suit numerical processing, and latent tensors are compact intermediate data that still require decoding.