Image
Image models
10 image models through Modelflare: GPT Images, Gemini native image, Wan, and Seedream, with the exact request contract and billing notes for each.
Image generation is not one contract. OpenAI-compatible models use /v1/images/*. Gemini image models use native generateContent. Use a Smart API Key or a key on the media group; see Authentication.
Your image request
OpenAI-compatible
POST /v1/images/generations GPT Images, Wan Image, and Seedream.
Gemini native
POST /v1/models/{model}:generateContent Text and reference images travel in the same contents[].parts[].
Media Skill
Let an agent choose an image model, estimate cost, and run the current request template.
Which endpoint to call
OpenAI Images vs Gemini native, and why mixing fields fails.
Every row links to the model page with the exact request contract. Prices and availability come from the main site.
OpenAI Images
Section titled “OpenAI Images”| Model | Endpoint | Billing | Pricing |
|---|---|---|---|
gpt-image-2.5-sunburst |
POST /v1/images/generations |
Per token | Pricing |
gpt-image-2.5-flare |
POST /v1/images/generations |
Per token | Pricing |
gpt-image-2 |
POST /v1/images/generations |
Per token | Pricing |
Gemini Image
Section titled “Gemini Image”| Model | Endpoint | Billing | Pricing |
|---|---|---|---|
gemini-3-pro-image |
POST /v1/models/{model}:generateContent |
Per token | Pricing |
gemini-3.1-flash-image |
POST /v1/models/{model}:generateContent |
Per token | Pricing |
gemini-3.1-flash-lite-image |
POST /v1/models/{model}:generateContent |
Per token | Pricing |
gemini-2.5-flash-image |
POST /v1/models/{model}:generateContent |
Per call | Pricing |
Wan / Seedream
Section titled “Wan / Seedream”| Model | Endpoint | Billing | Pricing |
|---|---|---|---|
wan2.7-image |
POST /v1/images/generations |
Per call | Pricing |
wan2.7-image-pro |
POST /v1/images/generations |
Per call | Pricing |
seedream-5-0-lite-260128 |
POST /v1/images/generations |
Per image | Pricing |