Meshy AI API Guide: How to Integrate AI 3D Generation Into Your App (2026)

Quick answer: Meshy AI API

The Meshy API is a REST API available on Studio plan ($60/month, 4,000 credits). Send POST requests to create 3D models from text or images. Receive model URLs via polling or webhooks. Export formats: FBX, OBJ, GLB, STL, USDZ. Full documentation at docs.meshy.ai. Rate limits scale with your plan.

Authentication

Authenticate with your API key:

• Get your API key: Log in to Meshy → Settings → API → Generate API Key.
• All requests require header: "Authorization: Bearer YOUR_API_KEY".
• API keys are plan-specific. Studio plan keys have 4,000 credits/month. Enterprise keys have custom limits.
• Keep your API key secret. Never expose it in client-side code. Use environment variables.
• Rate limits: 10 concurrent generations on Studio. 50 on Enterprise. Polling: 60 requests/minute.

Text to 3D via API

Generate 3D models from text prompts:

• Endpoint: POST https://api.meshy.ai/v2/text-to-3d
• Body: {"prompt": "a medieval sword with ornate hilt", "style": "realistic", "low_poly": false, "target_poly_count": 10000}
• Response: {"id": "task_abc123", "status": "PENDING"}
• Poll status: GET https://api.meshy.ai/v2/text-to-3d/task_abc123
• When status is "SUCCEEDED", response includes model_urls: {"fbx": "https://...", "obj": "https://...", "glb": "https://...", "stl": "https://..."}
• Cost: 1 credit per generation.
• Python example: requests.post("https://api.meshy.ai/v2/text-to-3d", json={"prompt": "..."}, headers={"Authorization": "Bearer KEY"})
• JavaScript example: fetch("https://api.meshy.ai/v2/text-to-3d", {method: "POST", headers: {"Authorization": "Bearer KEY", "Content-Type": "application/json"}, body: JSON.stringify({prompt: "..."})})

Image to 3D via API

Convert images to 3D models:

• Endpoint: POST https://api.meshy.ai/v2/image-to-3d
• Body: {"image_url": "https://example.com/product.jpg", "style": "realistic"}
• Alternatively, upload image: POST https://api.meshy.ai/v2/image-to-3d with multipart/form-data.
• Response: {"id": "task_def456", "status": "PENDING"}
• Poll for completion same as text-to-3D. Response includes model_urls.
• Cost: 1 credit per generation.
• Use case: E-commerce product 3D catalogs. Send product photos, get 3D models for AR.

AI Texturing via API

Apply PBR textures to existing models:

• Endpoint: POST https://api.meshy.ai/v2/texturing
• Body: {"model_url": "https://example.com/model.fbx", "prompt": "worn leather with brass studs", "quality": "hd"}
• Response: {"id": "task_ghi789", "status": "PENDING"}
• Poll for completion. Response includes textured_model_urls.
• Cost: 0.5 credits per texturing operation.
• Use case: Batch texture game assets or product models.

Webhook callbacks

Get notified when models are ready (no polling):

• Set webhook URL: POST https://api.meshy.ai/v2/webhooks {"url": "https://yourapp.com/webhook/meshy"}
• Meshy sends POST request to your URL when generation completes.
• Webhook payload: {"event": "task.completed", "task_id": "task_abc123", "status": "SUCCEEDED", "model_urls": {...}}
• Verify webhook signature: Meshy signs payloads with HMAC-SHA256. Verify with your webhook secret.
• Retry: Meshy retries failed webhooks 3 times with exponential backoff.
• Best practice: Use webhooks for production. Polling is fine for development.

Python code example — full pipeline

Complete Python example: text to 3D with download:

• import requests, time, urllib.request
• API_KEY = "your_api_key"
• HEADERS = {"Authorization": f"Bearer {API_KEY}", "Content-Type": "application/json"}
• # Create model
• resp = requests.post("https://api.meshy.ai/v2/text-to-3d", json={"prompt": "a fantasy sword", "style": "realistic"}, headers=HEADERS)
• task_id = resp.json()["id"]
• # Poll for completion
• while True:
• status = requests.get(f"https://api.meshy.ai/v2/text-to-3d/{task_id}", headers=HEADERS).json()
• if status["status"] == "SUCCEEDED": break
• time.sleep(5)
• # Download model
• urllib.request.urlretrieve(status["model_urls"]["fbx"], "sword.fbx")

JavaScript code example — full pipeline

Complete JavaScript/Node.js example:

• const API_KEY = process.env.MESHY_API_KEY;
• const headers = {"Authorization": `Bearer ${API_KEY}`, "Content-Type": "application/json"};
• // Create model
• const resp = await fetch("https://api.meshy.ai/v2/text-to-3d", {method: "POST", headers, body: JSON.stringify({prompt: "a fantasy sword", style: "realistic"})});
• const {id} = await resp.json();
• // Poll for completion
• async function poll() {
• const status = await fetch(`https://api.meshy.ai/v2/text-to-3d/${id}`, {headers}).then(r => r.json());
• if (status.status === "SUCCEEDED") return status;
• await new Promise(r => setTimeout(r, 5000));
• return poll();
• }
• const result = await poll();
• console.log("Download:", result.model_urls.fbx);

Production best practices

Scale your Meshy API integration:

• Use webhooks: Avoid polling in production. Set up webhook endpoints for real-time notifications.
• Queue management: Use a task queue (Redis, RabbitMQ) to manage generation requests. Avoid hitting rate limits.
• Error handling: Handle 429 (rate limit), 402 (insufficient credits), 500 (server error). Implement retry with backoff.
• Cache results: Store generated model URLs in your database. Avoid regenerating the same model.
• Credit monitoring: Check credit balance via GET /v2/credits. Alert when credits are low.
• Image optimization: For image-to-3D, compress and resize images before upload. Max 50MB.
• Async pipeline: Generate models asynchronously. Show users a "generating..." state. Notify on completion.
• Security: Never expose API keys in client-side code. Use a backend proxy. Validate webhook signatures.