Gorilla: Large Language Model Connected with Massive APIs

Gorilla is a state-of-the-art Large Language Model (LLM) designed to understand and accurately invoke over 1,600 APIs by interpreting natural language queries. Its primary aim is to reduce hallucination and offer a user-friendly experience for both developers and non-developers. Gorilla is built by fine-tuning LLaMA weights and harnesses the power of transformer architecture to provide semantically and syntactically correct API calls.

Key Features

• Accurately invoke 1,600+ APIs using natural language queries
• Reduce hallucination in LLMs
• User-friendly and adaptable to various needs and tools
• Open-source and constantly evolving with community contributions

Setup and Usage

To get started with Gorilla, follow these steps:

1. Visit the Gorilla GitHub repository and follow the instructions to set up your local environment.
2. Try Gorilla with the provided Google Colab notebook, or run it locally using the instructions in the inference folder.
3. Explore the APIBench dataset and evaluation code released by the Gorilla team.
4. Contribute your API to the growing APIZoo, following the contribution guide.

Integration with Other Tools

Gorilla is designed to work seamlessly with other LLM tools, such as Langchain, ToolFormer, and AutoGPT. Its adaptability makes it an ideal candidate for integration into a wide range of applications and toolchains.

System Requirements

To use Gorilla, you must have Python 3.10 or later installed. Earlier versions of Python will not compile. It is recommended to have a reliable internet connection and sufficient computational resources for optimal performance.

Community and Support

Gorilla is an open-source project, and the team behind it actively encourages community contributions and support. You can join the Gorilla Discord server to engage with the community, ask questions, or provide feedback. If you'd like to contribute to the project, follow the APIZoo contribution guide and be a part of the growing API ecosystem.