Skip to main content

Why Avala

What makes Avala different from Scale AI, Labelbox, Label Studio, and building your own pipeline.

Visualize

Multi-sensor MCAP/ROS playback with 8 panel types, GPU-accelerated 3D point cloud rendering across 6 visualization modes, and Gaussian Splat scene viewing — all in the browser with WebGPU.

Annotate

Professional annotation tools for bounding boxes, polygons, 3D cuboids, segmentation, polylines, keypoints, and classification — with quality control, object tracking, and managed labeling services.

Integrate

Python and TypeScript SDKs, REST API, MCP server for AI assistants, cloud storage connectors, webhooks, and model inference pipelines.

​
Explore the Platform

Visualization

Visualization

GPU-accelerated multi-sensor viewer with 8 panel types, 6 point cloud rendering modes, and Gaussian Splat support.

API Reference

API Reference

Programmatic access to datasets, projects, exports, and more.

Integrations

Integrations

Connect with S3, MCP, MCAP/ROS, webhooks, and inference pipelines.

​
Pick Your Language

Python

pip install avala

TypeScript

npm install @avala-ai/sdk

cURL

Direct REST API calls

Go

Use the REST API with Go

​
Quick Example

from avala import Client

client = Client(api_key="your-api-key")

# List your datasets
for ds in client.datasets.list():
    print(f"{ds.name} — {ds.item_count} items")

# Create an export
export = client.exports.create(project="your-project-uid")

​
Get Started

Quickstart

Create your first dataset and annotation project in under 60 seconds

Core Concepts

Understand datasets, projects, tasks, and the annotation lifecycle

Visualization

Explore multi-sensor recordings with GPU-accelerated viewers

​
Integrate with AI

MCP Server

Use Avala with Claude, Cursor, VS Code, and ChatGPT through the Model Context Protocol

Auto-Labeling

Connect SageMaker or custom models for AI-assisted annotation

​
Need Help?

Email support at support@avala.ai or open an issue on GitHub.