Image Generation

One request, one image URL back. Everything below was generated through these models:

Campaign direction 1 — glass bottle still life

Campaign direction 2 — palm shadow scene

Campaign direction 3 — beach product shot

Campaign direction 4 — botanical still life

POST /api/v1/images

{
  "model": "flux-2-klein",
  "prompt": "A golden retriever puppy in autumn leaves, soft lighting, photorealistic",
  "size": "1920x1920",
  "n": 1
}

Response

{
  "model": "flux-2-klein",
  "images": [{"url": "https://cdn.../generated-abc123.webp", "revised_prompt": "..."}],
  "cost": {"this_request": "$0.03", "per_image": "$0.03"},
  "metadata": {"latency_ms": 1847, "size": "1920x1920"}
}

Your image is at images[0].url — download it or hotlink it.

Parameters

Parameter	Default	Description
`prompt`	required	Be specific — subject, style, lighting, composition. Max 4,000 chars.
`model`	`nano-banana-2`	See the table below.
`size`	`1920x1920`	Exactly one of `1920x1920` (square), `2560x1440` (landscape), `1440x2560` (portrait).
`n`	`1`	1–4 images; each is billed.
`image`	—	Public HTTPS reference URL for image-to-image.
`aspect_ratio`	—	Aspect ratio for models that support it directly, e.g. `16:9`, `1:1` (Nano Banana, Imagen 4, FLUX.2 Flex).
`width`	—	Image width in pixels, 256–2048 (FLUX models only).
`height`	—	Image height in pixels, 256–2048 (FLUX models only).

size is a strict enum — 1024x1024 and other values are rejected with validation_error. Pick one of the three.

Which model?

Three answers cover most cases:

Drafting & volume

flux-2-klein — $0.03, under 2 seconds

The default

nano-banana-2 — $0.06, great text rendering

Maximum quality

flux-kontext-max — $0.10, premium editing

All 14 models — price, speed, specialty

Model	ID	Cost	Speed	Best for
GPT Image 2	`gpt-image-2`	$0.16	~10-20s	OpenAI quality, prompt adherence
FLUX Kontext Max	`flux-kontext-max`	$0.10	~5-15s	Premium editing, max fidelity
FLUX.2 Flex	`flux-2-flex`	$0.08	~5-10s	Maximum quality generation
FLUX.1 Pro Ultra	`flux-1-pro-ultra`	$0.08	~8-15s	Ultra-high resolution (4MP)
Recraft V3	`recraft-v3`	$0.08	~5s	Design, illustrations
Google Imagen 4	`google-imagen-4`	$0.08	~15s	Photorealism, text rendering
Nano Banana Pro	`nano-banana-pro`	$0.10	~5s	Professional assets
Seedream	`seedream`	$0.08	~8s	Realistic, artistic styles
FLUX.2 Pro	`flux-2-pro`	$0.05	~3-5s	Balanced quality/speed
FLUX Kontext Pro	`flux-kontext-pro`	$0.05	~5s	Image editing, style transfer
Nano Banana 2	`nano-banana-2`	$0.06	~3s	Best value, text rendering
FLUX.1 Fill	`flux-1-fill`	$0.07	~5s	Inpainting, outpainting
FLUX.2 Klein	`flux-2-klein`	$0.03	<2s	Fastest, batch work
Nano Banana	`nano-banana`	$0.03	~2s	Fast, budget-friendly

In code

import os, requests

def generate_image(prompt, model="flux-2-klein", size="1920x1920"):
    r = requests.post("https://www.ninjachat.ai/api/v1/images",
        headers={"Authorization": f"Bearer {os.environ['NINJACHAT_API_KEY']}"},
        json={"model": model, "prompt": prompt, "size": size})
    r.raise_for_status()
    return r.json()["images"][0]["url"]

draft = generate_image("A mountain landscape at sunset")                       # $0.03, <2s
final = generate_image("A mountain landscape at sunset, 8k, ultra detailed",
                       model="flux-2-flex", size="2560x1440")                  # $0.08

Try prompts live in the Playground — every run shows the exact request. Or let an agent drive: the MCP server gives Claude and Cursor these same models conversationally.

Get Started

Chat

Superpowers

Media & Search

Reference

Parameters

Which model?

Drafting & volume

The default

Maximum quality

In code

​Parameters

​Which model?

Drafting & volume

The default

Maximum quality

​In code

Parameters

Which model?

In code