Dioptra
p/dioptra-2
The metrics engine for better data and better models
Michael Seibel

LLM Sandbox by Dioptra — Everything you need to evaluate & improve prompts and LLMs

Featured
3
•
LLM Sandbox is an all-in-one environment to detect hallucinations, evaluate your prompts, select the right data to fine tune your models and track performance version after version.

Add a comment

Replies
Best
Farah Gasmi
Maker
đź“Ś
Hey PH, This is Farah! With @parnoux and @jacques_arnoux, we just launched LLM Sandbox as part of our open source project: katiML. With LLM Sandbox, teams can detect hallucinations, improve their prompts and chain of models. They also find the best data for fine-tuning faster. Would you use this? Why / why not? Thanks for the feedback! Links: Free trial: https://dev.dioptra.ai/register Our github https://github.com/dioptra-ai/ka... Documentation https://dioptra.gitbook.io/diopt... Slack community channel https://join.slack.com/t/dioptra...
Alexis Devenport
Congrats on the launch of LLM Sandbox, it sounds like a powerful tool for improving prompts and model chains. Looking forward to giving it a spin!
Neha
Congratulation on the launch LLM Sandbox!