# LLMReference

> LLMReference is a comprehensive database of AI language models. Compare models by benchmark scores, pricing, context length, capabilities, and provider — across 1,700+ models and 500+ model families.

## Sections

- [Models](https://www.llmreference.com/models) — Browse and compare 1,700+ AI language models with specs, benchmarks, and pricing
- [Model list for agents](https://www.llmreference.com/docs/models/llms.txt) — Agent-readable model list rendered from the current static seed data
- [Providers](https://www.llmreference.com/providers) — AI model providers including OpenAI, Anthropic, Google, Meta, Mistral, and more
- [Benchmarks](https://www.llmreference.com/benchmarks) — Benchmark scores and performance comparisons across models
- [Researchers](https://www.llmreference.com/researchers) — AI researchers and labs behind the models
- [Tools](https://www.llmreference.com/tools) — AI tools and coding assistants
- [Agents](https://www.llmreference.com/agents) — Agentic coding harnesses (Claude Code, Cursor, Aider, and the broader product directory)
- [Concepts](https://www.llmreference.com/concepts) — Core AI concepts: Model, Tools, Context, Harness, Agent, Agentic loop, MCP

## Quick Links

- Browse models: https://www.llmreference.com/models
- Agent-readable model list: https://www.llmreference.com/docs/models/llms.txt
- Browse providers: https://www.llmreference.com/providers
- Benchmarks: https://www.llmreference.com/benchmarks
- Agents directory: https://www.llmreference.com/agents
- Concepts glossary: https://www.llmreference.com/concepts

## Key Facts

- Coverage: 1,700+ models, 500+ model families, 130+ providers
- Data includes: context window, pricing (input/output per token), benchmark scores, release dates, license type
- No API required — all data is publicly accessible