Any LLM

Any LLM is a unified LLM management platform built by Mozilla.ai that lets developers bring their own API keys or use hosted models across providers like OpenAI, Anthropic, Mistral, Google Gemini, Meta, Hugging Face, and Perplexity. It provides a single API key that routes to multiple LLM providers, a real-time cost and usage dashboard, and client-side encryption so your keys never pass through Mozilla.ai's servers. Currently in beta and free to use, it is built on top of the open-source any-llm Python SDK for seamless provider switching without code changes.

• Client-side Encryption — API keys are encrypted with a key pair generated in your browser; only encrypted data is stored and keys never pass through Mozilla.ai's systems.
• Unified Cost Tracking — See real-time spending across all providers in one dashboard; identify expensive queries and compare costs across models.
• Single Key Simplicity — Get one API key configured to access multiple LLM providers, eliminating the need to manage dozens of keys in your backend.
• Provider Agnostic SDK — Install the any-llm-sdk Python library and switch providers without changing your application code.
• Unified Analytics — Track token usage, costs, and model performance across all your projects with detailed usage insights.
• Complete Privacy — Prompts and responses never reach Mozilla.ai's servers; only metadata (token counts, model names, timestamps) is collected.
• Multi-Tenant Support — Advanced multi-tenant configuration for teams managing multiple projects and clients.
• Budget System — Set and manage spending budgets per project or client to control LLM costs proactively.
• Self-Hosted Gateway — Deploy your own any-llm gateway for full infrastructure control.

To get started, sign up at any-llm.ai, create a project, add your provider API keys, and install the any-llm-sdk Python library to begin tracking usage in under 5 minutes.