Continuous visual context
The official model observes a sequence of frames as events unfold.
MOSS VL RealtimeBeta
MOSS VL Realtime: Streaming Video Understanding
Explore camera-based visual understanding with MOSS VL Realtime. Share live frames and ask questions with streaming text answers. Beta availability varies.
5 credits / minuteProvider unavailable
Camera connected
This Beta model is temporarily unavailable from MOSS. Its page and integration are ready; generation is disabled and no credits are charged.
Your camera is shared only while the session is running. Microphone audio is not sent.
Sign in to create1 credit per started 12 seconds; maximum 20 minutes per session.
Your finished audio, image or text will appear here.
01 / Inside the model
Understanding that keeps watching.
MOSS VL Realtime is designed for continuous visual perception. Unlike a one-off image question, its official model can keep observing a stream while generating an answer and update its response as the scene changes. Our integration is prepared for camera frames and text questions, with incremental text output. The upstream service is currently unavailable, so the camera session cannot start and credits are not charged.
Incremental text responses
The integration is designed to display answers as text arrives.
A different real-time tool
Use VL Realtime for visual questions; choose Live for a spoken conversation.
02 / Examples & listening room
Explore the official modelSee it. Hear it. Get a feel for it.
Official examples are credited to OpenMOSS. Studio samples were generated by us with the named API model. These are saved previews, not a result of your current settings. Research capabilities and available studio controls may differ.
Watching, reasoning and responding
Official MOSS-VL family video · original multilingual content03 / Use cases
Where it fits.
Following an activity
Explore visual changes and action sequences once the service becomes available.
Guided observation
Ask about what is visible while keeping a live camera context.
04 / Questions
A few things worth knowing.
Is MOSS VL Realtime available now?
No. Starting a session is disabled until the upstream service becomes available; no credits are charged.
How is the configured rate measured?
The configured rate is 5 credits per connected minute, billed in 12-second blocks, once the service is enabled.
How does it differ from MOSS Live?
VL Realtime is a streaming visual-understanding model with text output. MOSS Live supports real-time spoken interaction with optional visual context.
Made for your next idea
Choose a model, add your input and review the credit cost. Successful results stay in your account history; AI output may be inaccurate.
Text is rounded up to a whole credit. Audio is billed per started minute. Live voice uses one credit per 6 seconds.
Failed tasks are refunded automatically.
Result links can expire. Download files you want to keep.
Model documentation