ANT Documentation
ANT is the intelligent routing engine that sits between your application and LLM providers. It classifies prompts, optimizes tokens, and routes to the most cost-effective model for each request — powered by the Antbase platform.
Quick Start
Get up and running in 2 minutes
How It Works
Understand the ANT routing pipeline
Virtual Models
Use ant:auto, ant:fast, and more
Nests
Custom model pools with nest:name routing
Agents
Install and configure AI agents with ANT
Integration Guides
Python, TypeScript, LangChain, and more
API Reference
Complete endpoint documentation
Pricing
Plans, credits, and billing
ANT on ANTBASE
ANT
The intelligent routing engine. ANT analyzes each request and routes it to the optimal AI model — balancing quality, speed, and cost automatically. When you see “ANT selected model X” or “ANT routing”, it refers to the decision-making engine.
Antbase
The platform and ecosystem where ANT operates. Antbase connects ANT to the global AI ecosystem — providers, models, telemetry, and orchestration infrastructure. The API lives at antbase.ai.
Nest
A custom virtual model — a curated pool of models you hand-pick. Assign a nest to an API key and all requests route exclusively through your selected models. Use nest:your-name or assign via the dashboard. Great for compliance, budgets, or provider restrictions.
Example: “ANT analyzes each request and routes it to the optimal AI model using Antbase's global provider network. Nests let you restrict which models ANT can pick from.”
Key Features
Smart Routing
ANT classifies each prompt by task type and complexity, then routes to the optimal model based on your preferences.
Cost Optimization
Save up to 50% on LLM costs by using smaller models for simple tasks and reserving expensive models for complex work.
Drop-in Compatible
OpenAI-compatible API. Change your base URL and you're done. Works with any SDK or framework.