V-Modal AI provides Visual Memory capabilities for Mobile, Edge devices, Smart Glasses and Physical AI devices.
Find and retrieve any visual data (Video, Images) right in your app or your device.
by providing text or image query: "Find red car at night" or some input image.
No manual tags. No scrubbing timelines. Just ask.
Video is the fastest-growing data type on the planet — and the least searchable. Cameras, phones, smart glasses and physical-AI devices produce hours of footage that today can only be found by filename or hand-written tags.
V-Modal analyzes the actual frames and context of your videos, so people and machines can retrieve any moment by meaning — through natural language, an image, or another video. We ship it as small, typed SDKs that drop into the apps and devices developers already build.
Add an SDK to your app, configure a project with your API key, upload a video, and search your first moment — all in minutes.
- Get an API key: v-modal.com contact page
- Flutter docs: v-modal.github.io/vmodal_sdk_flutter
One platform, native toolkits everywhere. Mobile is live today — more surfaces are on the way.
| Platform | Status | Repo |
|---|---|---|
| 🐦 Flutter (Android + iOS) | ✅ Released | vmodal_sdk_flutter |
| 🤖 Android (Kotlin, native) | ✅ Released | vmodal_sdk_android |
| 🕶️ Smart Glasses / IoT / Camera | 🔜 Coming soon | — |
| 🌐 Web | 🔜 Coming soon | — |
- Natural-language video search — analyzes actual video frames and contextual data, so users search video content with conversational phrases.
- Multimodal queries — text-to-video, image-to-video and video-to-video search intents through one API.
- Cross-platform SDKs — a native Android Kotlin SDK and a cross-platform Flutter SDK with a uniform, typed surface.
- Built for mobile media — signed streaming URLs, chunked multipart uploads, progress and real cancellation, so massive video archives never crash or slow down devices.
- Media Asset Owner & Mobile Apps — let viewers jump to the precise scene inside video catalogs.
- Physical AI — give devices a visual memory they can query anytime.
- Smart Glasses — give devices a visual memory they can query anytime.
- E-commercee — visual search that surfaces product clips from an image or a description.
- CCTV — scan hours of footage for one specific visual indicator.
- 💬 Chat with the team on Discord
- 🌐 Developer hub: v-modal.com/developers
- ⭐ Star the SDKs to follow releases