Image Generation

Generate images from text prompts — or edit existing images — using models from OpenAI, Google Vertex AI, and more. The API is compatible with the OpenAI images endpoint, so existing integrations work without modification.

Generate an image

curl
Python

curl https://api.meshapi.ai/v1/images/generations \
  -H "Authorization: Bearer rsk_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-image-1-mini",
    "prompt": "A children drawing of a veterinarian and a baby otter",
    "response_format": "url"
  }'

from openai import OpenAI

client = OpenAI(api_key="rsk_...", base_url="https://api.meshapi.ai/v1")

response = client.images.generate(
    model="openai/gpt-image-1-mini",
    prompt="A children drawing of a veterinarian and a baby otter",
    response_format="url",
)
print(response.data[0].url)

Response:

{
  "created": 1717200000,
  "data": [
    {
      "url": "https://..."
    }
  ]
}

Request fields

Not all parameters are supported by every model. Support varies by provider — unsupported fields are silently ignored or return an error depending on the model. Check the model’s documentation or test with your target model before relying on a specific parameter.

Field	Type	Description
`model`	string	Image model ID (required)
`prompt`	string	Text description of the image to generate
`n`	integer	Number of images to generate (default: 1)
`size`	string	Image dimensions — e.g. `"1024x1024"`, `"1792x1024"`, or `"auto"` (default)
`quality`	string	`"auto"` (default), `"standard"` / `"hd"` (DALL-E 3), or `"low"` / `"medium"` / `"high"` (GPT image models)
`response_format`	string	`"url"` (default) or `"b64_json"`
`output_format`	string	Output encoding: `"png"`, `"jpeg"`, or `"webp"` — GPT image models only
`output_compression`	integer	Compression level 0–100 for `jpeg`/`webp` output — GPT image models only
`background`	string	Transparency: `"transparent"`, `"opaque"`, or `"auto"` — GPT image models only
`moderation`	string	Content-moderation level: `"low"` or `"auto"` — GPT image models only
`partial_images`	integer	Number of partial images streamed during generation (0–3) — streaming only

Edit an image

Transform an existing image — edit it with a prompt (optionally guided by a mask) or remove its background — by uploading it to POST /v1/images/edits. Unlike generation, this endpoint uses multipart/form-data (file upload), not JSON.

curl
Python

curl https://api.meshapi.ai/v1/images/edits \
  -H "Authorization: Bearer rsk_YOUR_KEY" \
  -F "model=openai/gpt-image-1" \
  -F "operation=edit" \
  -F "prompt=Add a party hat on the cat" \
  -F "image=@cat.png"

from openai import OpenAI

client = OpenAI(api_key="rsk_...", base_url="https://api.meshapi.ai/v1")

response = client.images.edit(
    model="openai/gpt-image-1",
    prompt="Add a party hat on the cat",
    image=open("cat.png", "rb"),
)
print(response.data[0].url)

Response:

{
  "created": 1717200000,
  "data": [
    {
      "url": "https://..."
    }
  ]
}

gpt-image-1 always returns the result as a base64 data URI in the url field. Pass response_format=b64_json for a b64_json field instead (supported on DALL·E models).

Operations

The operation field selects what to do. Support varies by provider:

Operation	Description	Providers
`edit` (default)	Edit guided by `prompt` (+ optional `mask` and reference images)	OpenAI, BytePlus
`remove_background`	Strip the background; returns a transparent image	OpenAI

Requesting an operation a provider doesn’t support returns 422. A provider with no image-edit support at all returns 501.

BytePlus supports edit only. remove_background is OpenAI-only.

Request fields

multipart/form-data:

Field	Type	Description
`model`	string	Image model ID (required)
`image`	file	Source image — PNG, JPEG, or WebP, up to 20 MB (required)
`prompt`	string	Edit instruction — required for `edit`
`operation`	string	`edit` (default) or `remove_background`
`mask`	file	Region to edit (OpenAI inpainting). Same formats as `image`
`reference_images`	file[]	Extra reference images (multi-image edit). 20 MB each, 28 MiB total
`n`	integer	Number of images to return (default: 1)
`size`	string	Output dimensions — e.g. `"1024x1024"` or `"auto"` (default)
`response_format`	string	`"url"` (default) or `"b64_json"` (DALL·E only)
`background`	string	`"transparent"`, `"opaque"`, or `"auto"`

The per-file caps sit under a hard 32 MiB ceiling on the whole request body, enforced at the edge — which is why the reference-image aggregate is 28 MiB rather than the sum of the per-file limits. An oversized request returns a plain HTML 413 page with no Mesh error envelope and no request ID. See Rate Limits & Spend Caps.

More examples

Remove a background:

curl https://api.meshapi.ai/v1/images/edits \
  -H "Authorization: Bearer rsk_YOUR_KEY" \
  -F "model=openai/gpt-image-1" \
  -F "operation=remove_background" \
  -F "image=@product.png"

Edit with multiple reference images (BytePlus Seedream):

curl https://api.meshapi.ai/v1/images/edits \
  -H "Authorization: Bearer rsk_YOUR_KEY" \
  -F "model=byteplus/seedream-4-5" \
  -F "operation=edit" \
  -F "prompt=Blend these into one product scene" \
  -F "image=@base.png" \
  -F "reference_images=@ref1.png" \
  -F "reference_images=@ref2.png"

Streaming

Some models support streaming image generation — partial image data is sent as the model renders it. Set "stream": true:

{
  "model": "openai/gpt-image-1-mini",
  "prompt": "...",
  "stream": true
}

The response is a text/event-stream. Its main job is to keep the connection open during a long generation: the server emits an initial processing chunk, then SSE comment pings until the image is ready. The stream always ends with data: [DONE].

Keep-alive chunk sequence

data: {"id":"img-...","object":"image.chunk","created":...,"model":"...","data":[],"status":"processing"}

: ping

: ping

data: {"id":"...","object":"image.chunk","created":...,"data":[{"url":"..."}]}

data: [DONE]

Available models

Model	Notes
`openai/gpt-image-1-mini`	Fast, cost-efficient — ideal for most use cases
`openai/gpt-image-1`	Highest quality, instruction-following
`openai/dall-e-3`	High quality, supports `hd` quality setting
`openai/dall-e-2`	Legacy, lower cost

Check GET /v1/models for the full live list of enabled image models and their pricing.

Introduction

Accounts & Access

Core Inference

Retrieval & Search

Multimodal

Operations

Mesh CLI

Platform

Reference

Generate an image

Request fields

Edit an image

Operations

Request fields

More examples

Streaming

Keep-alive chunk sequence

Available models

​Generate an image

​Request fields

​Edit an image

​Operations

​Request fields

​More examples

​Streaming

​Keep-alive chunk sequence

​Available models

Generate an image

Request fields

Edit an image

Operations

Request fields

More examples

Streaming

Keep-alive chunk sequence

Available models