Published by Video Runtime · Source check 2026-09-05 · Evidence method · Suggest a correction
EDITORIALLY VERIFIED PRODUCT INFORMATION
About Self Forcing
A research baseline for streaming video generation that trains models on their own autoregressive outputs. Simulates autoregressive inference during training and reuses attention state to reduce the mismatch between training and generation.
Key features
Publishes inference, training and a local interactive demo, providing a useful baseline for streaming-video research.
Getting started
Open the official repository, read its current README and licence, then follow the documented environment, checkpoint and inference steps. Start with the supplied example before using your own media or data.
Real-time interaction
Simulates autoregressive inference during training and reuses attention state to reduce the mismatch between training and generation.
Pricing
Apache-2.0; compute costs extra Repository code: Apache-2.0. Model weights and dependencies may have separate terms. No hosted price is estimated here. Budget for your own GPU or cloud usage after testing a representative workload.
Limitations
Authors specify Linux, at least 24 GB of NVIDIA GPU memory and 64 GB system RAM. This is not a hosted one-click consumer service.
EDITORIAL VIEW
Editor's Verdict
Consider Self Forcing if its documented workflow matches your experiment; start from the linked repository rather than an unofficial mirror.
Check the licence and run a small local evaluation before choosing it for production. No ratings or measured results are inferred from its publication.
How It Works
- Simulates autoregressive inference during training and reuses attention state to reduce the mismatch between training and generation.
What We Like
- Publishes inference, training and a local interactive demo, providing a useful baseline for streaming-video research.
Current Limitations
- Authors specify Linux, at least 24 GB of NVIDIA GPU memory and 64 GB system RAM. This is not a hosted one-click consumer service.
Pricing & Access
- Repository code: Apache-2.0. Model weights and dependencies may have separate terms.
- No hosted price is estimated here. Budget for your own GPU or cloud usage after testing a representative workload.
Verification Summary
- Official repository and licence reviewed on 2026-09-05. Video Runtime has not installed or benchmarked this project.
- This is a source review, not verification of demo uptime, output quality or performance on your hardware.
Technical notes & official performance
Official Performance Data
- The authors report real-time streaming on a single RTX 4090. This is an author claim, not a Video Runtime hands-on measurement.
Technical Notes
- Requires the documented Wan components and Self Forcing checkpoints. Lightweight decoding and lower precision can trade quality for speed.
Sources
- Official GitHub repository ↗github · accessed 2026-09-05
- Code licence ↗github · accessed 2026-09-05
- Official project documentation ↗official-docs · accessed 2026-09-05
Access, stage & mechanism
These are separate source-checked facts, not a live uptime monitor. Unverified fields are left open; transport and persistent sessions do not establish how frames are generated.
- Release stage
- Not verified
- Access mode
- Not verified
- Source availability
- Not verified
- Transport
- Not verified
- Interaction
- Not verified
- Session mode
- Not verified
- Generation mechanism
- Not verified
- Continuity method
- Not verified
- Licence status
- Not verified
- Tasks
- Not classified
CONTINUE EXPLORING
Related projects & models
Selected by research links and category, not by popularity.
A keyboard-controlled open-world research experience whose visuals and mechanics are generated by a world model.
Another frame streaming option in this section.Research demo, code, and model weights are linkedExplore long-running autoregressive video generation and the infrastructure behind interactive streams.
Another frame streaming option in this section.Public repository; self-hosted setup and compatible GPU requiredBuild interactive video streams with a diffusion system designed for configurable GPU pipelines.
Another frame streaming option in this section.Public repository; self-hosted setup and compatible GPU required
Run an action-conditioned interactive world model with open-ended rollout on a single desktop GPU.
Another frame streaming option in this section.Public repository; self-hosted installation and compatible GPU requiredRead how official-source and hands-on records stay separate in our Methodology.