FLUX.3 [dev] is coming with open weights for image, video, audio, and robot action prediction
multimodalart · x · 2026-07-24
The post says FLUX.3 [dev] is coming with open weights and support for image, video, audio, and robotic action prediction.
It also points to the model’s self-flow architecture, described as a new evolution of flow matching. The author praises the team for pushing architectural innovation in addition to scale.
Related event: Black Forest Labs Launches FLUX 3: A Unified Multimodal Foundation Model(26 posts)→
More from Multimodal
- Andrew Dai says visual reasoning is as fundamental to AGI as language — AndrewDai · 2026-07-24
- ComfyUI save-node plugin adds gallery, review triage, and workflow recovery — shootthesound · 2026-07-24
- Seedance 2.0 appears inside CapCut in a polished motion-art demo — bennash · 2026-07-24
- Midjourney’s `--draft --sref random` can produce 24 style directions for one job — ciguleva · 2026-07-24
- Debunked: Viral FLUX 3 Trailer is AI-Generated, Not an Official Release — gandamu_ml · 2026-07-24
- A ComfyUI node now auto-uploads generations to a shareable moodboard — massivebacon · 2026-07-24