Skip to content
Multilo Docs

Media & images

View images in a first-party viewer and let agents read images, audio, and video into context.

Research isn’t only text. Multilo views images natively and lets agents read images, audio, and video into context — so a figure, a recorded interview, or a lecture clip can be part of the work.

The Image Viewer

Open images in a first-party viewer built into the file explorer. Any image in your project opens cleanly, the same way documents do.

A scatter plot titled “Carbon burial recovers with restoration age” open in the Image Viewer, with zoom controls and the image size above it.
Images open in the built-in Image Viewer.

Agents see images

Agents can see the images you bring into context — a chart, a diagram, a scanned page — and reason about them, not just acknowledge that a file exists. Attach an image in chat and ask about it.

Audio & video

Multilo transcribes audio and video on-device so agents can read what was said — useful for interviews, talks, and recorded data. The same machinery powers Speech.

Local first

Image viewing and transcription run on your machine; media is sent to a model only when an agent genuinely needs it to answer.