Skip to content

REST API Reference

Ultralytics Platform provides a comprehensive REST API for programmatic access to datasets, models, training, and deployments.

Quick Start

# List your datasets
curl -H "Authorization: Bearer YOUR_API_KEY" \
  https://platform.ultralytics.com/api/datasets

# Run inference on a model
curl -X POST \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "file=@image.jpg" \
  https://platform.ultralytics.com/api/models/MODEL_ID/predict

Authentication

All API requests require authentication via API key.

Get API Key

  1. Go to Settings > API Keys
  2. Click Create Key
  3. Copy the generated key

See API Keys for detailed instructions.

Authorization Header

Include your API key in all requests:

Authorization: Bearer ul_your_api_key_here

Example

curl -H "Authorization: Bearer ul_abc123..." \
  https://platform.ultralytics.com/api/datasets

Base URL

All API endpoints use:

https://platform.ultralytics.com/api

Rate Limits

PlanRequests/MinuteRequests/Day
Free601,000
Pro30050,000
EnterpriseCustomCustom

Rate limit headers are included in responses:

X-RateLimit-Limit: 60
X-RateLimit-Remaining: 55
X-RateLimit-Reset: 1640000000

Response Format

All responses are JSON:

{
  "success": true,
  "data": { ... },
  "meta": {
    "page": 1,
    "limit": 20,
    "total": 100
  }
}

Error Responses

{
  "success": false,
  "error": {
    "code": "VALIDATION_ERROR",
    "message": "Invalid dataset ID",
    "details": { ... }
  }
}

Datasets API

List Datasets

GET /api/datasets

Query Parameters:

ParameterTypeDescription
pageintPage number (default: 1)
limitintItems per page (default: 20)
taskstringFilter by task type

Response:

{
    "success": true,
    "data": [
        {
            "id": "dataset_abc123",
            "name": "my-dataset",
            "slug": "my-dataset",
            "task": "detect",
            "imageCount": 1000,
            "classCount": 10,
            "visibility": "private",
            "createdAt": "2024-01-15T10:00:00Z"
        }
    ]
}

Get Dataset

GET /api/datasets/{datasetId}

Create Dataset

POST /api/datasets

Body:

{
    "name": "my-dataset",
    "task": "detect",
    "description": "A custom detection dataset"
}

Delete Dataset

DELETE /api/datasets/{datasetId}

Export Dataset

POST /api/datasets/{datasetId}/export

Returns NDJSON format download URL.

Projects API

List Projects

GET /api/projects

Get Project

GET /api/projects/{projectId}

Create Project

POST /api/projects

Body:

{
    "name": "my-project",
    "description": "Detection experiments"
}

Delete Project

DELETE /api/projects/{projectId}

Models API

List Models

GET /api/models

Query Parameters:

ParameterTypeDescription
projectIdstringFilter by project
taskstringFilter by task type

Get Model

GET /api/models/{modelId}

Upload Model

POST /api/models

Multipart Form:

FieldTypeDescription
filefileModel .pt file
projectIdstringTarget project
namestringModel name

Delete Model

DELETE /api/models/{modelId}

Download Model

GET /api/models/{modelId}/files

Returns signed download URLs for model files.

Run Inference

POST /api/models/{modelId}/predict

Multipart Form:

FieldTypeDescription
filefileImage file
conffloatConfidence threshold
ioufloatIoU threshold

Response:

{
    "success": true,
    "predictions": [
        {
            "class": "person",
            "confidence": 0.92,
            "box": { "x1": 100, "y1": 50, "x2": 300, "y2": 400 }
        }
    ]
}

Training API

Start Training

POST /api/training/start

Body:

{
    "modelId": "model_abc123",
    "datasetId": "dataset_xyz789",
    "epochs": 100,
    "imageSize": 640,
    "gpuType": "rtx-4090"
}

Get Training Status

GET /api/models/{modelId}/training

Cancel Training

DELETE /api/models/{modelId}/training

Deployments API

List Deployments

GET /api/deployments

Query Parameters:

ParameterTypeDescription
modelIdstringFilter by model

Create Deployment

POST /api/deployments

Body:

{
    "modelId": "model_abc123",
    "region": "us-central1",
    "minInstances": 0,
    "maxInstances": 10
}

Get Deployment

GET /api/deployments/{deploymentId}

Start Deployment

POST /api/deployments/{deploymentId}/start

Stop Deployment

POST /api/deployments/{deploymentId}/stop

Delete Deployment

DELETE /api/deployments/{deploymentId}

Get Metrics

GET /api/deployments/{deploymentId}/metrics

Get Logs

GET /api/deployments/{deploymentId}/logs

Query Parameters:

ParameterTypeDescription
severitystringINFO, WARNING, ERROR
limitintNumber of entries

Export API

List Exports

GET /api/exports

Create Export

POST /api/exports

Body:

{
    "modelId": "model_abc123",
    "format": "onnx"
}

Supported Formats:

onnx, torchscript, openvino, tensorrt, coreml, tflite, saved_model, graphdef, paddle, ncnn, edgetpu, tfjs, mnn, rknn, imx, axelera, executorch

Get Export Status

GET /api/exports/{exportId}

Activity API

Track and manage activity events for your account.

List Activity

GET /api/activity

Query Parameters:

ParameterTypeDescription
startDatestringFilter from date (ISO)
endDatestringFilter to date (ISO)
searchstringSearch in event messages

Mark Events Seen

POST /api/activity/mark-seen

Archive Events

POST /api/activity/archive

Trash API

Manage soft-deleted resources (30-day retention).

List Trash

GET /api/trash

Restore Item

POST /api/trash

Body:

{
    "itemId": "item_abc123",
    "type": "dataset"
}

Empty Trash

POST /api/trash/empty

Permanently deletes all items in trash.

Billing API

Manage credits and subscriptions.

Get Balance

GET /api/billing/balance

Response:

{
    "success": true,
    "data": {
        "cashBalance": 5000000,
        "creditBalance": 20000000,
        "reservedAmount": 0,
        "totalBalance": 25000000
    }
}

Micro-USD

All amounts are in micro-USD (1,000,000 = $1.00) for precise accounting.

Get Usage Summary

GET /api/billing/usage-summary

Returns plan details, limits, and usage metrics.

Create Checkout Session

POST /api/billing/checkout-session

Body:

{
    "amount": 25
}

Creates a Stripe checkout session for credit purchase ($5-$1000).

Create Subscription Checkout

POST /api/billing/subscription-checkout

Creates a Stripe checkout session for Pro subscription.

Create Portal Session

POST /api/billing/portal-session

Returns URL to Stripe billing portal for subscription management.

Get Payment History

GET /api/billing/payments

Returns list of payment transactions from Stripe.

Storage API

Get Storage Info

GET /api/storage

Response:

{
    "success": true,
    "data": {
        "used": 1073741824,
        "limit": 107374182400,
        "percentage": 1.0
    }
}

GDPR API

GDPR compliance endpoints for data export and deletion.

Export/Delete Account Data

POST /api/gdpr

Body:

{
    "action": "export"
}
ActionDescription
exportDownload all account data
deleteDelete account and all data

Irreversible Action

Account deletion is permanent and cannot be undone. All data, models, and deployments will be deleted.

API Keys API

List API Keys

GET /api/api-keys

Create API Key

POST /api/api-keys

Body:

{
    "name": "training-server",
    "scopes": ["training", "models"]
}

Delete API Key

DELETE /api/api-keys/{keyId}

Error Codes

CodeDescription
UNAUTHORIZEDInvalid or missing API key
FORBIDDENInsufficient permissions
NOT_FOUNDResource not found
VALIDATION_ERRORInvalid request data
RATE_LIMITEDToo many requests
INTERNAL_ERRORServer error

SDK Support

For easier integration, use the Ultralytics Python package.

Package Version Requirement

Platform integration requires ultralytics>=8.4.0. Lower versions will NOT work with Platform.

pip install "ultralytics>=8.4.0"

import os

from ultralytics import YOLO

# Set API key
os.environ["ULTRALYTICS_API_KEY"] = "ul_your_key"

# Train with Platform integration
model = YOLO("yolo11n.pt")
model.train(data="ul://username/datasets/my-dataset", project="username/my-project", name="experiment-1", epochs=100)

Webhooks

Webhooks notify your server of Platform events:

EventDescription
training.startedTraining job started
training.epochEpoch completed
training.completedTraining finished
training.failedTraining failed
export.completedExport ready

Webhook setup is available in Enterprise plans.

FAQ

How do I paginate large results?

Use page and limit parameters:

GET /api/datasets?page=2 &
limit=50

Can I use the API without an SDK?

Yes, all functionality is available via REST. The SDK is a convenience wrapper.

Are there API client libraries?

Currently, use the Ultralytics Python package or make direct HTTP requests. Official client libraries for other languages are planned.

How do I handle rate limits?

Implement exponential backoff:

import time


def api_request_with_retry(url, max_retries=3):
    for attempt in range(max_retries):
        response = requests.get(url)
        if response.status_code != 429:
            return response
        wait = 2**attempt
        time.sleep(wait)
    raise Exception("Rate limit exceeded")


📅 Created 0 days ago ✏️ Updated 0 days ago
glenn-jocher

Comments