Run Any LLM with a Single API: Introducing any-llm v1.0

Run any model like OpenAI, Claude, Mistral, or llama.cpp from one interface. any-llm v1.0 delivers production-ready stability, standardized reasoning output, and auto provider detection for seamless use across cloud and local models.

Nathan Brake

Nov 4, 2025 — 2 min read

One interface for every model, now production-ready

The LLM ecosystem is moving fast: new models, new providers, new API formats almost every month.

That’s why we built any-llm: a single, unified interface to use any model, cloud or local, without rewriting your stack every time. We wanted to make it easy for developers to use any large language model without being locked into a single provider.

Today, we’re thrilled to announce Version 1.0. A milestone that brings us closer to Mozilla.ai’s mission of building transparent, interoperable, and accessible AI.

Why any-llm

With any-llm, you can switch seamlessly between cloud and local model providers (like llamafile) using the same interface. This means less boilerplate code, fewer integration headaches, and more flexibility to pick what works best for you.

What’s New Since Launch

Since our initial release, we’ve greatly appreciated all of your feedback and contributions! Here’s what’s new and improved in any-llm v1.0:

Improved test coverage for stability and reliability
Responses API Support
List Models API to programmatically query supported models per provider
Re-usable client connections for improved performance.
Standardized reasoning output across all models, so you can confidently access LLM reasoning results, irrespective of the provider you’ve chosen.
Auto-updating provider compatibility matrix, showing which features are supported by which providers

Our guiding principle is simplicity: keep our interface stable, and let providers evolve behind the scenes by using the codebases that they already maintain.

Ready for Production

any-llm v1.0 is built for real workloads:

A stable, consistent API surface
Async-first APIs and re-usable client connections for high-throughput and streaming use cases.
Clear Deprecation and experimental notices to avoid surprises when API changes may occur.

Use any-llm to decouple your product logic from your model provider. Whether you're prototyping or scaling, your interface stays the same.

What’s Next

Version 1.0 isn’t the end! It’s the foundation for what’s next.

Coming soon:

Support for native batch completions
Adding support for new providers
Deeper integrations inside of other any-suite libraries: any-guardrail, any-agent, and mcpd

We’ve also got some other fun things in the works which we’ll be excited to share more about soon 😉