# cognee > cognee is an open-source AI memory engine that turns unstructured data into a queryable knowledge graph for agents and LLM applications. This file follows the llms.txt convention (https://llmstxt.org). It points language models at the most useful entry points on cognee.ai. ## Main pages - [Home](https://www.cognee.ai/): cognee — open-source AI memory engine for agents. - [Product](https://www.cognee.ai/product): How cognee turns unstructured data into a queryable memory graph for AI agents. - [Pricing](https://www.cognee.ai/pricing): Plans and pricing for cognee Cloud and self-hosted deployments. - [Consulting](https://www.cognee.ai/consulting): Hands-on engineering help to design, build, and ship AI memory systems with cognee. - [Research & evaluation results](https://www.cognee.ai/research-and-evaluation-results): Benchmarks and evaluation results for cognee against other memory and retrieval systems. - [About us](https://www.cognee.ai/about-us): The team and mission behind cognee. - [Blog](https://www.cognee.ai/blog): Deep dives, tutorials, and product news from the cognee team. - [Academy](https://www.cognee.ai/academy): Educational content on AI memory, knowledge graphs, and agentic systems. - [Wiki](https://www.cognee.ai/wiki): Reference material and concepts behind cognee. ## Developer resources - [Documentation (developer portal)](https://docs.cognee.ai): Full developer documentation, guides, and reference for building with cognee. - [Quickstart](https://docs.cognee.ai/getting-started/quickstart): Install cognee and build your first memory pipeline in minutes. - [REST API reference](https://docs.cognee.ai/api-reference/introduction): REST API endpoints, request/response schemas, and authentication (API keys and Bearer tokens). - [OpenAPI / Swagger spec](https://api.aws.cognee.ai/docs): Interactive OpenAPI (Swagger) explorer auto-generated from the cognee Cloud API specification. - [Python SDK reference](https://docs.cognee.ai/python-api): Reference for the cognee Python SDK. - [MCP server](https://docs.cognee.ai/cognee-mcp/mcp-overview): cognee MCP server: expose cognee memory to MCP-compatible agents, IDEs, and tools. - [cognee Cloud](https://docs.cognee.ai/cognee-cloud/overview): Managed cloud developer portal: API keys, dashboards, and hosted memory infrastructure. - [GitHub](https://github.com/topoteretes/cognee): Source code, issues, and examples for the open-source memory engine. ## Latest 10 blog posts - [Local AI Memory: Keeping Agent Memory Off the Cloud | Cognee](https://www.cognee.ai/local-ai-memory): What has to run on your own hardware for agent memory to count as local: four levels of locality, where cloud calls hide in extraction, embeddings, and reranking, how much hardware each stage needs, and a disconnect test for fully offline memory. - [AI Memory Tools vs. Databases: 5 Compared (2026) | Cognee](https://www.cognee.ai/ai-memory-tools-vs-databases): Compare LangMem, Mem0, cognee, Zep, and Letta by how much of the stack each AI memory tool takes over — from memory primitives over your own store to managed temporal graphs and memory built into the agent runtime. - [How to Evaluate AI Memory in 2026: 5 Tools Compared | Cognee](https://www.cognee.ai/how-to-evaluate-ai-memory-tools): Compare how Mem0, Zep, cognee, Letta, and LangMem let you evaluate AI memory — open benchmark harnesses, retrieval-versus-answer scoring, published results, and what you still have to test yourself. - [5 AI Memory Tools for Small Models and Local Agents | Cognee](https://www.cognee.ai/ai-memory-tools-for-small-models): Compare Graphiti, Mem0, cognee, Basic Memory, and Letta for small models and local agents — which AI memory tools need generative inference, what runs offline, and how much work lands on the agent’s own model. - [5 Open-Source LLM & Agent Evaluation Tools in 2026 | Cognee](https://www.cognee.ai/llm-agent-evaluation-tools): Compare DeepEval, Ragas, Promptfoo, Inspect AI, and Opik across output, retrieval, trajectory, and task-outcome evaluation layers — what each tool actually tests and how far it goes beyond scoring a single model response. - [5 New Databases for AI Workloads in 2026 | Cognee](https://www.cognee.ai/new-databases-for-ai-workloads): Compare HelixDB, turbopuffer, LanceDB, SurrealDB, and Turso across architecture, retrieval, and primary AI use — from graph-and-vector engines to object-storage-backed search and a database provisioned per agent. - [Keenable + cognee: Persistent Memory for Web Retrieval](https://www.cognee.ai/keenable-cognee-web-retrieval-memory): Keenable returns clean Markdown for any URL, and cognee turns it into persistent, graph-backed memory an agent can recall later — how the fetch backend, config, and pipeline fit together. - [5 Small Language Model Providers for Local and Specialized AI | Cognee](https://www.cognee.ai/small-language-model-providers): Compare Fastino, distil labs, Liquid AI, Arcee AI, and Mistral AI across model focus, deployment, and use case — from structured extraction and distilled students to on-device inference, sparse MoE, and general-purpose compact generation. - [Anthropic API Cost in 2026: Claude Pricing and Calculator | Cognee](https://www.cognee.ai/anthropic-api-cost): A breakdown of Anthropic API pricing for 2026: per-model Claude token rates, prompt caching and Batch API discounts, worked cost examples, and how much a memory layer can cut off the bill. - [Grok Pricing in 2026: API Costs and Calculator | Cognee](https://www.cognee.ai/grok-api-pricing): A breakdown of Grok API pricing for 2026: per-model token rates, what happens when a prompt crosses 200K tokens, caching and tool costs, and how much a memory layer can cut off the bill.