Decart's Lucy 2.5 Transforms Live Video at 30 FPS Without Drifting
Decart's Lucy 2.5 brings physically-aware VFX, Self-Anchoring for long-term edit stability, and sub-40ms latency to live 30 FPS video transformation
- Lucy 2.5 is live via the Decart API, delivering real-time video editing at 30 FPS with sub-40ms latency.
- Self-Anchoring is the headline innovation: the model uses its own generated output as a reference anchor, keeping edits stable for hours without drift.
- Physically-aware VFX (fire, water, slime, sand) now interact naturally with the scene environment via a massive synthetic training corpus.
- Inference stack uses MXFP8/NVFP4 quantization, dynamic sparse attention, and deeper kernel fusion for up to 4x speedups.
- Pricing is $0.02/sec for realtime and $0.04/sec for video edits at 720p; new accounts get free credits to start.
- Key use cases include live streaming VFX, virtual try-on, dynamic ad personalization, and robotics training data augmentation.
Lucy 2.5 is Decart's latest upgrade to its live video transformation model, and it pushes the boundary of what "real-time" means for generative video. It delivers higher-quality edits, physically-aware VFX, stronger temporal consistency, and even lower latency, all generated live at 30 FPS. This is not a clip generator. Every single frame is synthesized on the fly, responding to motion, lighting, and the scene as it evolves.
The Problem With Every Other Video AI
Most generative video tools work like a batch job: you submit a prompt, wait, and get a clip back. That model breaks the moment you need the output to respond to something happening right now. This represents a fundamental shift in how AI handles video: from static artifact generation to dynamic, mutable streaming. The deeper problem is that even systems that attempt continuous generation tend to drift. Small errors compound frame by frame until the subject's face morphs, the costume dissolves, or the scene collapses entirely.
Using Smart History Augmentation, the model can run continuously for hours without identity collapse or quality degradation, unlike competing models that typically maintain coherence for only 60 seconds. Lucy 2.5 extends this further with a new mechanism called Self-Anchoring, which is the headline technical innovation in this release.
Self-Anchoring: The Edit That Doesn't Drift
Self-Anchoring solves a fundamental problem in autoregressive generation (where each frame is generated conditioned on previous frames). Over time, the reference signal that guides the edit diverges from what the model is actually producing, causing the edit to slowly fade or mutate. Lucy 2.5 fixes this by having the model adopt a snapshot of its own generated output as the new reference anchor, rather than relying on the original input frame.
The key insight is not just using self-generated output at inference time, but training the model under those same self-anchored conditions. This means the model learns to actively exploit its own output as a guide, not just tolerate it. The results are significant:
- Edits persist over time: A character swap or costume change made in the first seconds of a stream stays locked in minutes later.
- Identity survives motion: Subjects can leave frame, turn around, and return without the edit dissolving.
- Stability compounds: Combined with Smart History Augmentation, streams hold identity, geometry, and texture over dramatically longer horizons.