List Image Models

GET /v1/images/models No API key needed View .md

The image models currently served, in OpenRouter's image model list shape, for images.listModels() in OpenRouter's SDKs. Each entry names the parameters the model accepts (supported_parameters: enum, range and boolean descriptors) and links its endpoint record in endpoints. Everything comes from the model's capability spec, so the limits listed are the limits enforced.

size is a constraint, not a list: a WxH string whose sides are multiples of multiple_of, whose area is between min_area and max_area pixels, and whose longer side is at most max_ratio times the shorter; where max_total_area is there, the n images of one request together may cover at most that many pixels (a model that renders them in one pass runs out of memory on more), counted as the engine renders them: on a size SeedVR2 enlarges after the engine, the render before it is enlarged. A range parameter that also takes values outside its range lists them as also (qwen-image-2.1's steps: 20 to 50, or 6). aspect_ratio and resolution are enums, and resolution.sizes is the table of the size each aspect ratio and resolution gives. Prices are on GET /v1/models (per megapixel).

Responses

application/json · image-models-response

1 field

data array of objects required
12 fields
- id string required
- name string required
- description string required
- created integer required
  
  Unix time in seconds.
- endpoints string required
  
  Relative URL of the model's endpoint record.
- architecture object required
  2 fields
  - input_modalities array of strings
  - output_modalities array of strings
- supported_parameters object required
  
  Every parameter the model accepts by name, including the ones that are also provider options.
- supports_streaming boolean required
- prompt_max_length integer or null
  
  The most characters prompt may have.
- inputs object or null
  
  The media a request may carry; null for a model that takes none (a request with an image or a mask is a 400).
  2 fields
  - reference object
    
    Reference images (input_references, image[] on /v1/images/edits, or images).
    
    5 fields
    
    image integer
    
    The most reference images a request may carry.
    
    formats array of strings
    
    The image types read as references, by their bytes.
    
    max_megapixels number or null
    
    The gateway scales a larger reference down to this many megapixels before the engine gets it; null for as sent.
    
    max_bytes integer
    
    The most bytes one image may have (a larger one is a 413).
    
    max_total_bytes integer
    
    The most bytes the images of a request may have together.
  - mask object
    
    The model takes a mask beside its references.
    
    1 field
    
    formats array of strings or null
- modes array of objects
  
  The model's modes in the order they are tried; the first whose when holds is the request's.
  6 fields
  - mode string
  - when object
    
    What selects the mode: inputs, media kinds that must all be present; parameters, values the request sent. Empty for the last mode, which takes the rest.
  - defaults object
    
    Parameter defaults the mode sets, by parameter name.
  - also object
    
    Values a parameter takes in this mode beside its range (minimax-h3: steps 8 with references, 4 without).
  - only object
    
    Values the mode holds a parameter to; anything else sent is a 400.
  - bill_multiplier object or null
    
    The megapixels a request in this mode bills are multiplied by base plus per_reference times the image references sent.
    
    2 fields
    
    base number
    
    per_reference number
- rules array of objects
  
  What a parameter value does to the others.
  3 fields
  - when object
    
    The parameter values that bring the rule in.
  - require object
    
    Values another parameter must then take; an unsent one takes the first.
  - refuse array of strings
    
    Parameters the request may not send then.

curl https://api.vultrinference.com/v1/images/models

import requests

response = requests.get(
    "https://api.vultrinference.com/v1/images/models",
)
print(response.json())

const response = await fetch("https://api.vultrinference.com/v1/images/models");
console.log(await response.json());

Response · 200

{
    "data": [
        {
            "prompt_max_length": 4000,
            "inputs": {
                "reference": {},
                "mask": {}
            },
            "modes": [
                {}
            ],
            "rules": [
                {}
            ]
        }
    ]
}