Librechat
Enhanced ChatGPT Clone: Features Agents, MCP, DeepSeek, Anthropic, AWS, OpenAI, Responses API, Azure, Groq, o1, GPT-5, Mistral, OpenRouter, Vertex AI, Gemini, Artifacts, AI model switching, message search, Code Interpreter, langchain, DALL-E-3, OpenAPI Actions, Functions, Secure Multi-User Auth, Pre
Last verified:
What is Librechat?
LibreChat is a free, open-source AI platform that brings together all your AI conversations in one unified, customizable interface. It integrates the best language models from every major provider—including OpenAI, Anthropic, Google, Azure, AWS Bedrock, and dozens more—into a single interface without vendor lock-in or subscriptions. The platform is MIT licensed, allowing users to freely use, modify, and distribute the software.
Key features include advanced agents with file handling, code interpretation, and API actions; a secure multi-language code interpreter that executes code with zero setup; multimodal chat for image and file analysis; Model Context Protocol (MCP) support to connect to any tool or service; persistent context across conversations; built-in web search with reranking for live internet access; generative UI with Code Artifacts for creating React, HTML, and Mermaid diagrams; conversation and message search; prompt templates; presets for saving predefined settings; and enterprise-ready SSO with OAuth, SAML, LDAP, and two-factor authentication. The interface is available in 30+ languages with community-driven translations.
LibreChat is designed for developers, power users, enterprises, startups, universities, and government agencies who want full control over their AI infrastructure. It is ideal for users who want to avoid subscription costs, bypass rate limits, consolidate multiple AI providers into one interface, self-host for privacy and compliance, or customize their AI workflow with custom endpoints and plugins. With 36k GitHub stars, 31.3M Docker pulls, and 335 contributors, it is trusted by companies worldwide.
Librechat pricing
Pricing model: Freemium
LibreChat is completely free and open-source under the MIT License with no subscriptions or restrictions. You can use, modify, and distribute it freely. The core platform is free, but you must pay for AI API usage separately when using providers like OpenAI, Anthropic, or Google—this is pay-as-you-go token-based pricing. There is no official hosted cloud version; you self-host on your own infrastructure. Third-party managed hosting is available through services like Elestio (hourly pricing based on CPU/RAM/disk resources). Some third-party resellers offer hosted plans (Hobby $9.99/month, Enthusiast $24.99/month, Pro $49.99/month) but these are not official LibreChat pricing.
Librechat pros
- Fully open-source and MIT licensed with no subscriptions
- Supports 10+ AI providers including OpenAI, Anthropic, Google, Azure, AWS Bedrock
- Self-hosted deployment keeps your data private and compliant
- No vendor lock-in—use any model provider you choose
- Advanced agents with file handling, code interpretation, and API actions
- Secure multi-language code interpreter with zero setup
- Model Context Protocol (MCP) support for custom tool integration
- Multimodal chat supports image and file analysis
- Generative UI with Code Artifacts for React, HTML, Mermaid diagrams
- Built-in web search with reranking for live internet access
- Enterprise-ready SSO with OAuth, SAML, LDAP, and 2FA
- Presets feature to save and share conversation settings as JSON
- 30+ language support with community-driven translations
- Conversation and message search for instant file/code snippet lookup
- Persistent context across conversations so AI remembers you
- Custom endpoints via librechat.yaml for any OpenAI API-compatible service
- 36k GitHub stars and 31.3M Docker pulls prove community trust
- Prompt templates and plugins extend functionality
Librechat cons
- Requires self-hosting technical knowledge for setup and maintenance
- Default MongoDB backend uses SSPL license (concern for commercial SaaS)
- You must pay for AI API usage separately (pay-as-you-go model)
- No official hosted cloud version—fully self-hosted only
- Configuration involves multiple files (librechat.yaml, .env, docker-compose)
- Community-maintained tool may have less uptime guarantee than commercial services
- API usage costs can escalate quickly without monitoring
- YAML configuration syntax errors can break custom endpoint setup
- Mongo diversity requires FerretDB alternative for full MIT compliance
Frequently asked questions about Librechat
What is LibreChat?
LibreChat is a free, open-source AI platform that brings together all your AI conversations in one unified, customizable interface. It integrates language models from every major provider—OpenAI, Anthropic, Google, Azure, AWS Bedrock, and dozens more—into a single interface without vendor lock-in or subscriptions. It is MIT licensed and self-hostable.
Is LibreChat free to use?
Yes, the LibreChat platform itself is completely free and open-source under the MIT License with no subscriptions or restrictions. However, you must pay for AI API usage separately when using providers like OpenAI, Anthropic, or Google. The API pricing is token-based and pay-as-you-go.
Can I self-host LibreChat?
Yes, LibreChat is designed for self-hosting on your own infrastructure. You can deploy it using Docker with docker-compose, and your data stays on your server unless you configure external services. The code interpreter runs in isolated sandboxes for security. This gives you full privacy and compliance control.
What AI models and providers does LibreChat support?
LibreChat supports OpenAI, Anthropic, Google, Azure, AWS Bedrock, and dozens more providers. It works with any OpenAI API-compatible service as a custom endpoint. Popular models include GPT-3.5-turbo, GPT-4, Claude 3.7 Sonnet, and reasoning models like o1 and o3-mini.
How do I add custom AI endpoints like OpenRouter or Ollama?
Add custom endpoints by creating or editing the librechat.yaml configuration file in your project root. Define your endpoint (name, API URL, models, display settings) in librechat.yaml, store API keys in the .env file using ${VAR_NAME} syntax, and mount the config via docker-compose.override.yml for Docker deployments. Restart LibreChat after changes.
What is the presets feature in LibreChat?
Presets let you save and load predefined settings for conversations. You can choose the endpoint, model, custom instructions, temperature, Top P, frequency penalty, and presence penalty. Presets can be imported/exported as JSON files, set as default, and shared with others on Discord. Load a preset automatically for new conversations by setting it as default.
Does LibreChat support MCP (Model Context Protocol)?
Yes, LibreChat supports Model Context Protocol (MCP) to connect to any tool or service. You can define custom MCP tools in the tools section of librechat.yaml and reference them in agents. This allows you to tailor LibreChat to your exact workflow with custom integrations.
What enterprise security features does LibreChat offer?
LibreChat offers enterprise-ready SSO with OAuth, SAML, and LDAP integration, two-factor authentication, rate limiting, and moderation tools. The platform is self-hosted so your data stays on your infrastructure for full privacy and compliance. File upload rate limiting is configurable through librechat.yaml.
Can I create custom AI agents without coding?
Yes, LibreChat supports no-code agents with file handling, code interpretation, and API actions. You can create custom AI assistants for specific tasks through the interface without writing code. Agents integrate with MCP tools and can be configured via the librechat.yaml file.
How do I get help or contribute to LibreChat?
You can ask questions, share ideas, and report bugs on GitHub Discussions. Chat with the community in real-time on Discord. Follow @LibreChatAI on X (Twitter) for updates and news. Watch tutorials and demos on YouTube. Contribute code, documentation, translations, and ideas on GitHub. The project has 335 contributors and a dedicated maintenance team.