Inference

Create response

Generate an image or video from a text prompt, or transform an existing image. Built on the Open Responses spec, a provider-agnostic API standard.

Text → image and image → image: Use FLUX models (black-forest-labs/FLUX.2-dev)
Text → video and image → video: Use Wan models (Wan-AI/Wan2.2-T2V-A14B-Diffusers, Wan-AI/Wan2.2-I2V-A14B-Diffusers)

Pass a plain string as input for text-to-image/video. Pass a structured content array for image-to-image/video.

POST

responses

import os
import base64
from openai import OpenAI

client = OpenAI(
    base_url="https://api.modular.com/v1",
    api_key=os.environ["MODULAR_API_KEY"],
)

response = client.responses.create(
    model="black-forest-labs/FLUX.2-dev",
    input="A serene mountain landscape at sunset",
    extra_body={
        "provider_options": {
            "image": {"height": 1024, "width": 1024, "steps": 28}
        }
    },
)

image_data = response.output[0].content[0].image_data
with open("output.png", "wb") as f:
    f.write(base64.b64decode(image_data))

{
  "id": "<string>",
  "object": "<string>",
  "model": "<string>",
  "output": [
    {
      "content": [
        {
          "type": "<string>",
          "image_data": "<string>",
          "text": "<string>"
        }
      ]
    }
  ]
}

Authorizations

Authorization

string

header

required

Modular Cloud API token. Obtain from the API Tokens page.

Body

application/json

model

string

required

Model identifier. See the supported models page.

Example:

"black-forest-labs/FLUX.2-dev"

input

required

Model input. Pass a plain string for text-to-image/video, or a structured content array for image-to-image/video.

seed

integer

Random seed for reproducible outputs.

provider_options

object

Show child attributes

Response

200 - application/json

Response object containing generated content.

string

required

object

string

required

model

string

required

output

object[]

required

Show child attributes

Create deploymentCreate a new dedicated inference endpoint. `name` and `model_id` are immutable after creation. > **Not yet available.** Self-serve deployment creation is coming soon. [Sign up for private preview access](https://console.modular.com/signup) to be notified when it launches.

import os
import base64
from openai import OpenAI

client = OpenAI(
    base_url="https://api.modular.com/v1",
    api_key=os.environ["MODULAR_API_KEY"],
)

response = client.responses.create(
    model="black-forest-labs/FLUX.2-dev",
    input="A serene mountain landscape at sunset",
    extra_body={
        "provider_options": {
            "image": {"height": 1024, "width": 1024, "steps": 28}
        }
    },
)

image_data = response.output[0].content[0].image_data
with open("output.png", "wb") as f:
    f.write(base64.b64decode(image_data))

{
  "id": "<string>",
  "object": "<string>",
  "model": "<string>",
  "output": [
    {
      "content": [
        {
          "type": "<string>",
          "image_data": "<string>",
          "text": "<string>"
        }
      ]
    }
  ]
}