World Model Index
Viva
Prompt-directed video and audio with image inputs and configurable chunks.
- Provider
- Sequencer
- Category
- Continuous video
- Published price
- $40.00 /hrActive stream time
- Resolution
- 1344 × 768
- Frame rate
- 24 fps output
- Access
- Hosted API · request access
Capabilities
- Live prompts
- Image input
- Frame reset
- Variable chunks
- Audience voting
Pricing & billing
$40.00/hr of active stream time, as listed on the Worldstream API page. Confirm the current session quote with Sequencer.
Billing basis: Active stream time
Quality evidence
No verified Elo score for this model.
Video Elo measures clip preference. World consistency and interaction latency require separate tests.
Related comparisons
Notes
Viva is Sequencer’s distilled H3 adaptation and streaming framework. The provider technology specification lists 4-step inference, 1344 × 768 video and 24 fps output.
The API documents WebRTC delivery, changing the prompt, resetting the start frame and continuous last-frame handoff. Audience voting and story-state management belong to the hosted Worldstreams experience.
Configurable chunk sizes are provider-confirmed; the range is unpublished. The tutorial uses 15-second scene handoffs.
24 fps describes output playback. Independent latency, throughput, and cross-model quality testing are pending.
Sources
Last checked October 15, 2026. Specifications and prices use provider sources; Elo links to the evaluator.
- Worldstream APIhttps://sequencer.media/worldstream-api
- Worldstream workflowhttps://sequencer.media/posts/tutorials/minimax-h3-realtime-worldstreams
- Worldstreamshttps://worldstreams.ai/
- Viva technologyhttps://sequencer.media/demo
Compare models
View pricing and capabilities side by side.