Encyclopedia Evalica / Datasets / Multimodal dataset
Multimodal dataset
/multimodal 'day.tuh.seht/A dataset containing non-text inputs/outputs (images, audio, PDFs), requiring multimodal evals and storage. Multimodal datasets often need specialized rubrics and display tooling for review. (noun)
“Our multimodal dataset checks whether the model reads values from screenshots.”
Customer example
Navan built a multimodal dataset of raw audio recordings (not just transcripts) to evaluate its voice agent, so scoring captures tone and nuance alongside content. Read more
Related Datasets terms
From the docs
Braintrust is the observability platform for agents in production. By actively applying intelligence to agent traces and automatically surfacing critical patterns, Braintrust helps teams at Notion, Stripe, Box, OpenAI, and Cloudflare ship quality agents at scale.
Start building