S
SeedEdit Developer API
Advertisement
⚡ GPU Cloud Partner: Need dedicated multi-region cluster endpoints? Deploy on Leonardo Enterprise.
REST & Client-Side Endpoints

SeedEdit API Documentation

Integrate SeedEdit causal diffusion, instruction-based image edits, and deterministic seed generation into your web apps, mobile apps, and automated creative pipelines.

Free Client REST

Pollinations Flux Engine

$0 cost, zero API keys required. Direct client-side GET requests with instant HTTPS response stream.

Cloud GPU Managed

Segmind SeedEdit 3.0 API

Dedicated ByteDance research model checkpoint hosted on NVIDIA H100 with sub-2s inference latency.

Open Source Checkpoint

Replicate / Cog Container

Self-hostable or serverless container deployment via PyTorch and Diffusers pipelines.

GET https://image.pollinations.ai/prompt/{prompt}
$0 Server Fees

The primary engine powering SeedEdit.app. You can embed and call this URL directly inside any HTML <img src="..."> tag or via JavaScript fetch().

Query Parameters

Parameter Type Default Description
prompt string (URL Encoded) Required Text prompt or editing instruction to synthesize.
seed integer Random Deterministic 32-bit random seed (0 to 2,147,483,647).
width integer 1024 Output pixel width (e.g. 1024, 1280, 720).
height integer 1024 Output pixel height (e.g. 1024, 720, 1280).
model string flux Diffusion model checkpoint: flux or turbo.
nologo boolean true Suppresses artificial watermark overlay.
curl -X GET "https://image.pollinations.ai/prompt/Cyberpunk%20geisha%20portrait%2C%208k?seed=9482014&width=1024&height=1024&nologo=true&model=flux" \
  --output seededit_output.jpg
POST https://api.segmind.com/v1/seededit-3.0
Production Latency: ~1.8s

For commercial high-volume workloads requiring ByteDance SeedEdit 3.0 image-to-image instruction editing with API SLAs.

import requests
import base64

api_key = "SG_YOUR_SEGMIND_KEY"
url = "https://api.segmind.com/v1/seededit-3.0"

# Read base image as Base64
with open("input_portrait.jpg", "rb") as f:
    image_b64 = base64.b64encode(f.read()).decode("utf-8")

payload = {
    "image": image_b64,
    "instruction": "Make background a rainy Tokyo alley with neon reflections",
    "seed": 9482014,
    "cfg_scale": 7.5,
    "denoise_strength": 0.7
}

response = requests.post(url, json=payload, headers={"x-api-key": api_key})

with open("seededit_edited.jpg", "wb") as out:
    out.write(response.content)
print("Edit successfully generated with SeedEdit 3.0!")

Interactive API Request Tester

Live In-Browser
Live API Response Waiting for trigger
{ "status": "idle", "message": "Click 'Send Test Request' to simulate endpoint call" }