Terra Cotta

Terracotta is a user-friendly platform for experimenting with Large Language Models (LLMs) efficiently. It allows managing multiple fine-tuned models in one place, making it easy to iterate and improve them quickly. The tool offers qualitative and quantitative evaluation features, enabling users to compare model outputs using metrics like accuracy and BLEU. Secure data uploading is a crucial feature for the fine-tuning process. Users can store their data on the platform and use it to fine-tune models for classification and text generation tasks with just a few clicks. Terracotta seamlessly integrates with popular LLM providers like OpenAI and Cohere, providing flexibility to work with different providers. Created by LLM experts Beri Kohen and Lucas Pauker, both Stanford AI graduates, Terracotta streamlines the LLM development workflow, making it more accessible. Stay updated on their latest developments by joining their email list.