# Classer > AI text classification API. Classify and tag text with a single API call — no training required. From $0.08/1M tokens, under 200ms latency. ## Why Classer over a general-purpose LLM? Classer is not a wrapper around OpenAI or Anthropic. It runs a 30B-parameter frontier LLM optimized for classification on dedicated GPU infrastructure. - **Speed** — Under 200ms P95. General-purpose LLMs are moving toward reasoning models (5s+); classification doesn't need chain-of-thought. - **Cost** — $0.08–$0.60/1M tokens + 10M free tokens/month. Caching makes repeated classifications free. - **Zero prompt engineering** — No prompts to craft, no output parsing, no format inconsistencies. Structured JSON every time. - **No model management** — Models upgrade without breaking your integration. No deprecation notices. - **3 lines of code** — Pass text + labels, get a result. Best for teams that need classification as a feature, not a project. ## Docs - [Introduction](https://docs.classer.ai/introduction): What Classer is, code examples, use cases - [Quickstart](https://docs.classer.ai/quickstart): Get up and running in 5 minutes - [Authentication](https://docs.classer.ai/authentication): API keys and auth setup - [Pricing & Limits](https://docs.classer.ai/pricing): Speed tiers, rate limits, free tier, validation limits - [API Overview](https://docs.classer.ai/api-reference/overview): All endpoints, request/response format, error handling - [Classify endpoint](https://docs.classer.ai/api-reference/classify): POST /v1/classify — single-label classification - [Tag endpoint](https://docs.classer.ai/api-reference/tag): POST /v1/tag — multi-label tagging - [Fast Batch](https://docs.classer.ai/api-reference/batch): Async batch processing for large datasets - [Python SDK](https://docs.classer.ai/sdks/python): pip install classer - [TypeScript SDK](https://docs.classer.ai/sdks/typescript): npm install classer-ai ## Full context - [llms-full.txt](https://docs.classer.ai/llms-full.txt): Complete API reference for AI consumption