> ## Documentation Index
> Fetch the complete documentation index at: https://docs.orisu.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Model settings

> Reference for the settings generation models expose, grouped by what they control: format, prompt behavior, generation strength, video, upscale, speech, music, and text.

This page describes the settings you'll encounter across Orisu's generation models. They're grouped by what they control.

<Note>
  Which settings appear, their allowed values, and their defaults vary by model. A node's config panel is the source of truth for the model you've selected. Not every model exposes every setting below, and some are only available on specific tiers.
</Note>

## Format and composition

Controls for the shape, size, and encoding of the output.

<ParamField body="aspect_ratio" type="string">
  Width-to-height ratio of the output, such as `1:1` or `16:9`.
</ParamField>

<ParamField body="resolution" type="string">
  Output resolution, often expressed as a preset like `720p` or `1080p`.
</ParamField>

<ParamField body="target_aspect_ratio" type="string">
  For aspect-ratio changes, the ratio the input is reframed to.
</ParamField>

<ParamField body="target_resolution" type="string">
  For upscaling, the resolution to produce, such as `4k`.
</ParamField>

<ParamField body="output_format" type="string">
  File format of the output, such as `png` or `jpeg`.
</ParamField>

<ParamField body="quality" type="string">
  Quality preset that trades render time and cost against fidelity.
</ParamField>

<ParamField body="background" type="string">
  Whether the output background is opaque or transparent.
</ParamField>

<ParamField body="input_fidelity" type="string">
  How closely an edit preserves the original input image.
</ParamField>

## Prompt controls

Settings that change how the model interprets and expands your prompt.

<ParamField body="negative_prompt" type="string">
  Describes what to avoid in the output.
</ParamField>

<ParamField body="enhance_prompt" type="boolean">
  Lets the model rewrite your prompt for better results before generating.
</ParamField>

<ParamField body="enable_prompt_expansion" type="boolean">
  Expands a short prompt into a more detailed one automatically.
</ParamField>

<ParamField body="expand_prompt" type="boolean">
  Alternate name some models use for automatic prompt expansion.
</ParamField>

<ParamField body="prompt_optimizer" type="boolean">
  Optimizes the prompt for the model's preferred phrasing.
</ParamField>

<ParamField body="enable_web_search" type="boolean">
  Allows the model to search the web for current context while generating.
</ParamField>

## Generation strength (image and video)

Controls for how literally the model follows the prompt and how much it transforms the input.

<ParamField body="guidance_scale" type="number">
  How strictly the model adheres to the prompt. Higher values follow it more closely.
</ParamField>

<ParamField body="cfg_scale" type="number">
  Classifier-free guidance scale. Another name for prompt adherence strength on some models.
</ParamField>

<ParamField body="num_inference_steps" type="number">
  Number of denoising steps. More steps can add detail at the cost of speed.
</ParamField>

<ParamField body="strength" type="number">
  How much the output departs from the input image. Higher values change it more.
</ParamField>

<ParamField body="reference_strength" type="number">
  How strongly reference images influence the result.
</ParamField>

<ParamField body="style_strength" type="number">
  How strongly a style reference is applied.
</ParamField>

<ParamField body="remove_color_bias" type="boolean">
  Neutralizes color casts carried over from reference inputs.
</ParamField>

<ParamField body="rendering_speed" type="string">
  Speed preset that trades render time against quality.
</ParamField>

## Video

Settings specific to video generation and editing.

<ParamField body="duration" type="number">
  Length of the generated clip, in seconds. Each model supports a specific set
  of lengths. In the studio this is a slider that snaps to those values, and
  runs triggered through the API or MCP adjust an unsupported length to the
  nearest one the model accepts. Models with a fixed clip length, or that take
  their length from the source video, don't expose this setting.
</ParamField>

<ParamField body="generate_audio" type="boolean">
  Generates a matching audio track alongside the video.
</ParamField>

<ParamField body="generate_background_music" type="boolean">
  Adds generated background music to the clip.
</ParamField>

<ParamField body="camera_fixed" type="boolean">
  Locks the camera in place so the shot doesn't move.
</ParamField>

<ParamField body="movement_amplitude" type="string">
  How much motion the model introduces into the scene.
</ParamField>

<ParamField body="keep_audio" type="boolean">
  Retains the source clip's audio when editing video.
</ParamField>

<ParamField body="keep_original_sound" type="boolean">
  Preserves the original sound rather than replacing it.
</ParamField>

<ParamField body="character_orientation" type="string">
  How a character is framed or facing in the output.
</ParamField>

<ParamField body="preserve_audio" type="boolean">
  Keeps the existing audio track when upscaling video.
</ParamField>

<ParamField body="acceleration" type="string">
  Speed-versus-quality preset for the render.
</ParamField>

<ParamField body="output_quality" type="string">
  Quality preset for the produced video.
</ParamField>

## Upscale (image)

Image upscalers expose detail controls that shape how much the model invents versus preserves. The most common ones:

<ParamField body="scale" type="number">
  Upscale factor, such as `2` or `4`.
</ParamField>

<ParamField body="creativity" type="number">
  How much new detail the model may add. Higher values invent more.
</ParamField>

<ParamField body="resemblance" type="number">
  How closely the result stays true to the original image.
</ParamField>

<ParamField body="fractality" type="number">
  How much fine, repeating texture detail is introduced.
</ParamField>

<ParamField body="hdr" type="number">
  Strength of high-dynamic-range enhancement.
</ParamField>

<ParamField body="sharpen" type="number">
  Amount of sharpening applied to the result.
</ParamField>

<ParamField body="face_enhancement" type="boolean">
  Applies dedicated restoration to faces. Some models add `face_enhancement_strength` and `face_enhancement_creativity` to tune it.
</ParamField>

<ParamField body="optimized_for" type="string">
  Tunes the upscaler for a content type, such as photos or art.
</ParamField>

<ParamField body="engine" type="string">
  Selects the underlying upscaling engine when a model offers more than one.
</ParamField>

<ParamField body="mode" type="string">
  Upscaling mode the model runs in.
</ParamField>

<Note>
  Upscalers vary widely. Beyond these, individual models expose their own detail knobs; the config panel lists every control available for the model you pick.
</Note>

## Speech (text-to-speech)

Settings that shape synthesized voice.

<ParamField body="stability" type="number">
  Consistency of the voice. Higher values are steadier; lower values are more expressive.
</ParamField>

<ParamField body="similarity_boost" type="number">
  How closely the output matches the source voice.
</ParamField>

<ParamField body="style" type="number">
  How much stylistic exaggeration is applied to the delivery.
</ParamField>

<ParamField body="speed" type="number">
  Speaking rate of the voice.
</ParamField>

## Music

Settings for music generation.

<ParamField body="duration" type="number">
  Length of the generated track, in seconds.
</ParamField>

<ParamField body="force_instrumental" type="boolean">
  Generates an instrumental track with no vocals.
</ParamField>

## Text and LLMs

Settings for text-generation models.

<ParamField body="reasoning_effort" type="string">
  How much the model reasons before answering. Higher effort can improve quality at the cost of speed.
</ParamField>

<ParamField body="enable_web_search" type="boolean">
  Allows the model to search the web for current information.
</ParamField>

<ParamField body="output_schema_preset" type="string">
  Constrains the output to a structured format using a predefined schema.
</ParamField>