Black Forest Labs teases FLUX.3 as an open-weights multimodal backbone
multimodalart · x · 2026-07-24
Black Forest Labs’ FLUX.3 [dev] is teased as an open-weights multimodal backbone for content creation and action prediction.
According to the post and the accompanying slide, the model is positioned for image, video, audio, and robotic action prediction, with the author highlighting the self-flow architecture as a notable evolution of flow matching. The message is that BFL is continuing to innovate at the architecture level, not just by scaling model size.
More from Multimodal
- Midjourney style code share: --sref 2912175708 — tisch_eins · 2026-09-11
- Astra storyboards plus Minimax H3 per-shot generation boost video success rates — Hailuo_AI · 2026-09-11
- MiniMax H3 MAX nails cooking anime clips: 15-second curry demo with prompts shared — Hailuo_AI · 2026-09-11
- MiniMax Music Production Toolkit 2.5 for ComfyUI adds full mastering chain — Vivid_Promise1700 · 2026-09-11
- New Node Finder for ComfyUI ranks fresh nodes by star velocity and recency — Luke2642 · 2026-09-11
- Using a finisher move on one mosquito with MiniMax H3 MAX — the bug survives — Hailuo_AI · 2026-09-11