Black Forest Labs says it is training FLUX 3 on images, video and audio together
Start here
Black Forest Labs researchers Andreas Blattmann and Dustin Podell say the company is training FLUX 3 jointly on still images, video and audio, arguing that a still shows where an object is while video shows how it got there. Blattmann, a co-inventor of latent diffusion, says ordinary 2D video carries evidence about how objects look from other angles, and that BFL derives its video upscaler from the same FLUX 3 backbone so detail stays temporally consistent. Podell describes an early Kontext experiment in which a few hours of training on a small labeled photo set let the model place a logo on a shirt, and notes that users restoring degraded family photos with FLUX was an unanticipated use. The claims come from the researchers' own account in a Latent Space interview; no benchmark results or independent evaluation were reported.
Generated from reports · updated 23 min ago
Report timeline
Follow the reports to see the event from different sides.
- Latent SpaceBlack Forest Labs on building visual intelligence and training FLUX 3 on image, video and audio
In a Latent Space interview, Black Forest Labs' Andreas Blattmann and Dustin Podell describe training FLUX 3 on still images, video and audio together, arguing that a still shows where an object is while video shows how it got there. Andreas, co-inventor of latent diffusion, says ordinary 2D video carries evidence about how objects look from other angles, and BFL derives its video upscaler from the same FLUX 3 backbone so detail stays temporally consistent. Dustin recounts an early Kontext experiment where a few hours of training on a small labeled photo set let the model place a logo on a shirt, and notes users restoring degraded family photos with FLUX was an unanticipated use.
Heat for this event
Not enough continuous observations yet to draw a trend.