LLMs have seen an impressive wave of advances, 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 a challenging task even for today's state-of-the-art LLMs, largely due to their inability to generate accurate input arguments, and their tendency to hallucinate the wrong usage of an API call. We release Gorilla LLM that surpasses the performance of all close-sourced and open-sourced LLMs on 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 in LLMs. Gorilla is an open-source project having served hundreds of thousand user requests and an energetic community supporting it. Check out the project at https://gorilla.cs.berkeley.edu/.
Shishir G. Patil is a CS PhD student at UC Berkeley, where he is advised by Prof. Joseph Gonzalez and Prof. Prabal Dutta. He is interested in designing and building efficient machine-learning systems for the two extremes of computing - edge and multi-cloud. Recently, he is focused on teaching LLMs to use tools through API calls. His works include Gorilla LLM, Skyplane, and POET. He was a Research Fellow at Microsoft Research, and has interned at Amazon Science, Apple, and Google Brain.
Come connect with the global community of thinkers and disruptors who are building and deploying the next generation of AI and ML applications.