jimeng

Jimeng - Image Generation API

Generate images from text prompts using ByteDance Jimeng (Seedream 4.5), a high-quality image generation model.

Requires Rebyte API auth — $AUTH_TOKEN and $API_URL are set up per the agent's system prompt; use them as Bearer token and base URL.

Text-to-Image Generation

Create an image from a text description.

curl -X POST "$API_URL/api/data/jimeng/generate" \
  -H "Authorization: Bearer $AUTH_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "A futuristic cityscape at sunset with flying cars",
    "size": "2K",
    "aspectRatio": "16:9"
  }'

---

Image-to-Image Generation

Provide a reference image URL to guide generation.

curl -X POST "$API_URL/api/data/jimeng/generate" \
  -H "Authorization: Bearer $AUTH_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "Transform this into a cyberpunk style",
    "image": "https://example.com/photo.jpg"
  }'

Note: The image parameter must be a publicly accessible URL (not base64).

---

Parameters

Name	Type	Required	Default	Description
prompt	string	Yes	-	Text description of the image to generate
image	string	No	-	Public URL of a reference image (for image-to-image)
model	string	No	seedream-4.5	Model variant
size	string	No	2K	Output size: 1K, 2K, or 4K
aspectRatio	string	No	1:1	Output aspect ratio

Aspect Ratios:

Ratio	Use Case	2K Resolution
1:1	Square (social media, icons)	2048x2048
16:9	Landscape (presentations, banners)	2560x1440
9:16	Portrait (mobile, stories)	1440x2560
4:3	Standard landscape	2304x1728
3:4	Standard portrait	1728x2304
3:2	Photo landscape	2304x1536
2:3	Photo portrait	1536x2304
21:9	Ultra-wide (cinematic)	2560x1097

---

Response

{
  "image": {
    "base64": "iVBORw0KGgoAAAANSUhEUgAA...",
    "mimeType": "image/png",
    "dataUrl": "data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAA..."
  },
  "description": null
}

Field	Description
image.base64	Base64-encoded image data
image.mimeType	Image MIME type
image.dataUrl	Ready-to-use data URL for HTML/CSS
description	Always null (Seedream does not return descriptions)

---

Using with Python

import subprocess
import requests
import base64
import json
from pathlib import Path

# Get auth token and API URL
AUTH_TOKEN = subprocess.check_output(["/home/user/.local/bin/rebyte-auth"]).decode().strip()
with open('/home/user/.rebyte.ai/auth.json') as f:
    API_URL = json.load(f)['sandbox']['relay_url']

HEADERS = {"Authorization": f"Bearer {AUTH_TOKEN}"}

def generate_image(
    prompt: str,
    image_url: str = None,
    size: str = "2K",
    aspect_ratio: str = "1:1"
) -> dict:
    """Generate an image from text, or use a reference image."""
    payload = {
        "prompt": prompt,
        "size": size,
        "aspectRatio": aspect_ratio
    }

    if image_url:
        payload["image"] = image_url

    response = requests.post(
        f"{API_URL}/api/data/jimeng/generate",
        headers=HEADERS,
        json=payload
    )
    return response.json()

def save_image(result: dict, filepath: str) -> None:
    """Save generated image to a file."""
    if "image" in result:
        image_data = base64.b64decode(result["image"]["base64"])
        Path(filepath).write_bytes(image_data)
        print(f"Saved to {filepath}")
    else:
        print(f"Error: {result.get('error', 'Unknown error')}")

# Example 1: Text-to-image
result = generate_image(
    prompt="A serene mountain landscape at dawn with mist in the valley",
    aspect_ratio="16:9"
)
save_image(result, "landscape.png")

# Example 2: Image-to-image with reference URL
result = generate_image(
    prompt="Transform this into a watercolor painting",
    image_url="https://example.com/photo.jpg",
    size="2K"
)
save_image(result, "watercolor.png")

---

Using with Node.js

const fs = require('fs');
const { execSync } = require('child_process');

// Get auth token and API URL
const AUTH_TOKEN = execSync('/home/user/.local/bin/rebyte-auth').toString().trim();
const authConfig = JSON.parse(fs.readFileSync('/home/user/.rebyte.ai/auth.json'));
const API_URL = authConfig.sandbox.relay_url;

async function generateImage(prompt, options = {}) {
  const payload = {
    prompt,
    size: options.size || '2K',
    aspectRatio: options.aspectRatio || '1:1'
  };

  if (options.imageUrl) {
    payload.image = options.imageUrl;
  }

  const response = await fetch(`${API_URL}/api/data/jimeng/generate`, {
    method: 'POST',
    headers: {
      'Authorization': `Bearer ${AUTH_TOKEN}`,
      'Content-Type': 'application/json'
    },
    body: JSON.stringify(payload)
  });
  return response.json();
}

function saveImage(result, filepath) {
  if (result.image) {
    const buffer = Buffer.from(result.image.base64, 'base64');
    fs.writeFileSync(filepath, buffer);
    console.log(`Saved to ${filepath}`);
  } else {
    console.error('Error:', result.error || 'Unknown error');
  }
}

// Example: Text-to-image
(async () => {
  const result = await generateImage(
    'A neon-lit cyberpunk street scene at night',
    { aspectRatio: '21:9', size: '2K' }
  );
  saveImage(result, 'cyberpunk.png');
})();

---

Prompt Tips

Bad:  "A dog"
Good: "A golden retriever puppy playing in autumn leaves, warm afternoon sunlight, shallow depth of field"

• Be specific: describe subject, setting, lighting, mood, style
• Seedream excels at photorealistic and artistic styles
• Chinese prompts work well (the model is trained on Chinese + English)

---

Error Handling

Missing or Invalid Auth Token:

{
  "error": "Missing sandbox token"
}

Solution: Run rebyte-auth and include the token in your request.

API Error:

{
  "error": "Request failed",
  "message": "Jimeng API error (400): ..."
}

Check your prompt and parameters. Some prompts may be blocked by content safety filters.

---

Delivering Output

After generating images, upload them to the Artifact Store so the user can access them.

Important Notes

• Default size is 2K (higher quality than nano-banana's default 1K)
• Generated images include watermarks
• The image parameter for image-to-image must be a public URL, not base64
• Chinese prompts are well supported
• Content safety filters may block certain prompts