streamingSending a model's response token by token as it is generated, so the reader sees text immediately instead of waiting for the whole answer.Full definition → generated video changes when moderation can intervene: viewers see frames before the complete video exists. For its upcoming real-time video models, Runway describes checking inputs first, inspecting frames during generation and stopping the stream when prohibited content appears. It then checks the completed video before allowing downloads or sharing, because a sequence can reveal a problem that individual frames do not. Runway reports average safety-review latency below half a second, but also reports brief exposure to prohibited material in some tests. The system therefore limits exposure after detection while retaining a separate whole-video review. Moderation design.
Terms in this piece · Glossary
streaming — Sending a model's response token by token as it is generated, so the reader sees text immediately instead of waiting for the whole answer.
LoRA — A cheap way to fine-tune a model by training a small add-on layer instead of changing all of the model's weights.
mixture-of-experts — A model built from many specialist sub-networks where only a few activate per token, giving big-model capability at small-model running cost.
