---
title: "Create image"
method: POST
path: "/v1/images/generations"
tags: ["Images"]
---

# Create image

`POST /v1/images/generations`

## Request

- Base URL: `https://api.together.ai`
- URL: `https://api.together.ai/v1/images/generations`
- Auth: HTTP bearer

Use an image model to generate an image for a given prompt.

## Request body

- object
  - `prompt` string, required: A description of the desired images. Maximum length varies by model.
  - `model` union, required: The model to use for image generation.<br> <br> [See all of Together AI's image models](https://docs.together.ai/docs/serverless-models#image-models)
    - 'black-forest-labs/FLUX.1-schnell-Free' | 'black-forest-labs/FLUX.1-schnell' | 'black-forest-labs/FLUX.1.1-pro'
    - string
  - `steps` integer: Number of generation steps.
  - `image_url` string: URL of an image to use for image models that support it.
  - `seed` integer: Seed used for generation. Can be used to reproduce image generations.
  - `n` integer: Number of image results to generate.
  - `height` integer: Height of the image to generate in number of pixels.
  - `width` integer: Width of the image to generate in number of pixels.
  - `negative_prompt` string: The prompt or prompts not to guide the image generation.
  - `response_format` 'base64' | 'url': Format of the image response. Can be either a base64 string or a URL.
  - `guidance_scale` number: Adjusts the alignment of the generated image with the input prompt. Higher values (e.g., 8-10) make the output more faithful to the prompt, while lower values (e.g., 1-5) encourage more creative freedom.
  - `output_format` 'jpeg' | 'png': The format of the image response. Can be either be `jpeg` or `png`. Defaults to `jpeg`.
  - `image_loras` object[]: An array of objects that define LoRAs (Low-Rank Adaptations) to influence the generated image.
    - `path` string, required: The URL of the LoRA to apply (e.g. https://huggingface.co/strangerzonehf/Flux-Midjourney-Mix2-LoRA).
    - `scale` number, required: The strength of the LoRA's influence. Most LoRA's recommend a value of 1.
  - `reference_images` string[]: An array of image URLs that guide the overall appearance and style of the generated image. These reference images influence the visual characteristics consistently across the generation.
  - `disable_safety_checker` boolean: If true, disables the safety checker for image generation.

## Response `200`

Image generated successfully

- ImageResponse
  - `id` string, required
  - `model` string, required
  - `object` 'list', required: The object type, which is always `list`.
  - `data` union[], required
    - union
      - ImageResponseDataB64
        - `index` integer, required
        - `b64_json` string, required
        - `type` 'b64_json', required
      - ImageResponseDataUrl
        - `index` integer, required
        - `url` string, required
        - `type` 'url', required

## Changes

- **2026-10-22** `bb5257b3173a`: 1 info
  - endpoint added

[Operation history](https://skmtc.dev/together/apis/together-apis/changes/v1/images/generations/post.md)

---

[API](https://skmtc.dev/together/apis/together-apis.md) - [All operations](https://skmtc.dev/together/apis/together-apis/llms.txt) - [OpenAPI document](https://skmtc.dev/together/apis/together-apis/revisions/93438ddac711?raw)