AISuite is a lightweight Python library from Andrew Ng’s team that gives you one simple, unified interface to many generative AI providers. It offers a single Chat Completions-style API across providers such as OpenAI, Anthropic, Google and others, so you can switch models by changing a model string like “openai:gpt-…” to “anthropic:claude-…” instead of rewriting client code for every vendor SDK. A second layer adds tool use, letting models call functions through the same unified interface. AISuite has more than 16,000 stars on GitHub and was still actively developed in 2026.
Anyone who has built with several LLM providers knows the friction: different SDKs, different request shapes, different error handling. AISuite removes it by normalizing chat completions behind one client and one convention for model names, while still letting you pass provider-specific options when you need them. That makes it easy to compare providers on the same prompts, fall back when one service is down, or migrate an application to a better model without a rewrite. The tool-calling layer extends the same simplicity to agents, so a function defined once can be offered to models from different providers.
AISuite is free and open source under the MIT license. You pay only for the model API usage billed by whichever providers you connect.
AISuite is for Python developers who work with more than one LLM provider and want to compare, switch or combine models without maintaining a separate integration for each vendor.