Supported Models
JourneyAPI supports 30 image generation models across 9 families. All models are accessible via POST /imagine — just set the model field. Midjourney is the default. Most non-Midjourney models are synchronous: the task resolves immediately without polling.
How to Choose
Fastest / cheapest
flux-2-dev (1cr)flux-schnell (1cr)nano-banana (2cr)
Good for drafts, testing, and high-volume pipelines.
Best quality
midjourney (4cr)grok-image-2 (4cr)flux-2-max (4cr)
Top-tier photorealism and creative output.
Edit / modify an image
flux-kontext-pro (4cr)seededit (3cr)ideogram-v3-edit (6cr)
Requires image_url. Edits subject while preserving identity.
Text in images
ideogram-v3 (6cr)nano-banana-2 (4cr)gpt-image-1 (2cr)
Best legibility of typography, labels, and signs.
Full Model List
| Model ID | Provider | Type | Credits | Sync? |
|---|---|---|---|---|
| midjourney | Midjourney | T2I (grid) | 4 | Poll |
| flux-pro | Black Forest Labs | T2I | 4 | Poll |
| flux-dev | Black Forest Labs | T2I | 2 | Poll |
| flux-schnell | Black Forest Labs | T2I | 1 | Poll |
| nano-banana | T2I | 2 | Yes | |
| nano-banana-2 | T2I / I2I | 4 | Yes | |
| nano-banana-hd | T2I | 3 | Yes | |
| seedream-3 | ByteDance | T2I | 2 | Yes |
| seedream-4 | ByteDance | T2I | 3 | Yes |
| seedream-4.5 | ByteDance | T2I | 3 | Yes |
| seedream-5-lite | ByteDance | T2I | 3 | Yes |
| seededit | ByteDance | Edit | 3 | Yes |
| gpt-image-1 | OpenAI | T2I / Edit | 2 | Yes |
| gpt-image-1.5 | OpenAI | T2I / Edit | 3 | Yes |
| gpt-4o-image | OpenAI (proxy) | T2I / Edit | 2 | Yes |
| grok-image | xAI | T2I / I2I | 3 | Yes |
| grok-image-2 | xAI | T2I / I2I | 4 | Yes |
| z-image | Alibaba | T2I | 2 | Yes |
| qwen-image | Alibaba | T2I | 4 | Yes |
| qwen-image-edit | Alibaba | Edit | 4 | Yes |
| ideogram-v3 | Ideogram | T2I | 6 | Yes |
| ideogram-v3-edit | Ideogram | Edit | 6 | Yes |
| ideogram-v3-remix | Ideogram | Edit | 6 | Yes |
| ideogram-v3-reframe | Ideogram | Reframe | 6 | Yes |
| recraft | Recraft | T2I | 3 | Yes |
| flux-2-dev | Black Forest Labs | T2I | 1 | Yes |
| flux-2-flex | Black Forest Labs | T2I | 2 | Yes |
| flux-2-max | Black Forest Labs | T2I | 4 | Yes |
| flux-kontext-pro | Black Forest Labs | Edit | 4 | Yes |
| flux-kontext-max | Black Forest Labs | Edit | 8 | Yes |
Response Shape
Midjourney — 2×2 grid
Midjourney is the only model that returns a grid of four images. The response includes bothoriginal_image_url (the full 2×2 grid) and image_urls (array of 4 individual images, 0-indexed).
{
"status": "finished",
"original_image_url": "https://cdn.midjourney.com/abc/grid_0.png",
"image_urls": [
"https://cdn.midjourney.com/abc/0_0.png",
"https://cdn.midjourney.com/abc/0_1.png",
"https://cdn.midjourney.com/abc/0_2.png",
"https://cdn.midjourney.com/abc/0_3.png"
]
}All other models — single image
Every non-Midjourney model returns a single image in theimage_url field.image_urls andoriginal_image_url will be absent.
{
"status": "finished",
"image_url": "https://cdn.example.com/generated.png"
}Edit Models — Requires image_url
The following models are image editors: they transform a source image according to your prompt. You must provide image_url pointing to the source image. Without it, the request will either fail validation or produce a plain text-to-image result that ignores your editing intent.
seededitqwen-image-editideogram-v3-editideogram-v3-remixideogram-v3-reframeflux-kontext-proflux-kontext-max