Skip to content

Research repository on interfacing LLMs with Database APIs. Inspired by the Berkeley Gorilla LLM.

Notifications You must be signed in to change notification settings

weaviate/DBGorilla

Repository files navigation

Weaviate Gorilla

The Weaviate Gorilla project is a collection of research experiments in two core concepts:

(1) Agents and Function Calling with Weaviate

(2) Creating synthetic training and testing data with Generative Feedback Loops

Weaviate Gorilla Leaderboard

Natural Language Commands to Weaviate Queries

Illustrated below, the Weaviate Gorilla translates natural language commands into Weaviate queries. We separate Weaviate queries into 3 categories of difficulty (simple, moderate, complex) based on how many unique query operators they require.

NL Command to API Visual

News

🎙️ Shishir Patil and Tianjun Zhang on the Weaviate Podcast - link

🎥 Fine-tuning LLMs to use Weaviate's GraphQL APIs on Weaviate Youtube - link

📝 Fine-tuning LLMs to use Weaviate's GraphQL APIs on the Weaviate Blog - link

🎥 Gorilla LLM Explained by Connor Shorten on Weaviate YouTube - link

🎥 SQL-PaLM Explained by Connor Shorten on Weaviate YouTube - link

Weaviate Gorilla

About

Research repository on interfacing LLMs with Database APIs. Inspired by the Berkeley Gorilla LLM.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published