Gorilla: Writing API calls with LLMs

Gorilla: Writing API calls with LLMs

Introduction

Gorilla is a groundbreaking Large Language Model (LLM) that is revolutionizing the way we interact with APIs. Developed by a team of researchers at UC Berkeley and Microsoft Research, Gorilla is trained on three massive machine learning hub datasets: Torch Hub, TensorFlow Hub, and HuggingFace. The team is also rapidly adding new domains, including Kubernetes, GCP, AWS, OpenAPI, and more.

What Makes Gorilla Unique?

Gorilla's unique selling point is its ability to provide appropriate API calls. In a zero-shot setting, Gorilla outperforms GPT-4, Chat-GPT, and Claude. It is extremely reliable and significantly reduces hallucination errors, a common issue with other LLMs.

Accessibility and Commercial Use

Gorilla is designed to be user-friendly. You can try Gorilla in just 60 seconds without any sign-ups or installations, thanks to its integration with Google Colab. Moreover, with Apache 2.0 licensed LLM models, you can use Gorilla commercially without any obligations. The team behind Gorilla is excited to hear your feedback and welcomes API contributions as they continue to build this open-source project.

Gorilla's Potential and Performance

Large Language Models have seen an impressive wave of advances recently, with models now excelling in a variety of tasks such as mathematical reasoning and program synthesis. However, their potential to effectively use tools via API calls remains unfulfilled. This is where Gorilla shines.

Gorilla, a fine-tuned LLaMA-based model, surpasses the performance of GPT-4 in writing API calls. When combined with a document retriever, Gorilla demonstrates a strong capability to adapt to test-time document changes, enabling flexible API updates and version changes.

Gorilla also substantially mitigates the issue of hallucination, commonly encountered when prompting LLMs directly. To evaluate the model's ability, the team introduced APIBench, a comprehensive dataset consisting of HuggingFace, TorchHub, and TensorHub APIs. The successful integration of the retrieval system with Gorilla demonstrates the potential for LLMs to use tools more accurately, keep up with frequently updated documentation, and consequently increase the reliability and applicability of their outputs.

Conclusion

Gorilla represents a significant step forward in the field of Large Language Models and their interaction with APIs. By providing accurate API calls and reducing hallucination errors, Gorilla is set to become a valuable tool for developers and businesses alike. The open-source nature of the project also encourages collaboration and continuous improvement, making it a promising development in the world of AI and machine learning.

For more information, you can check out Github or read the research paper detailing its development and capabilities.

This blog post is based on the information available on the Gorilla project page. For the most accurate and up-to-date information, please refer to the original source.