FLUX 3 expands into one multimodal backbone for image, video, audio and action
pmttyji · reddit · 2026-07-24
FLUX 3 is introduced as a single multimodal backbone for image, video, audio, and action prediction.
The launch plan says the team will roll out capabilities over the coming weeks and months after an early-access phase for feedback and safety testing. Planned releases include:
- FLUX 3 Video for video and audio generation/editing via APIs and private weights
- FLUX-mimic and FLUX 3 Action for action prediction with research and commercial partners
- FLUX 3 Image for image synthesis and editing
- FLUX 3 Dev, an open-weight multimodal backbone for content creation and action prediction
The post also says more technical details about the underlying approach will be shared later.
Related event: BFL Launches FLUX 3: A Unified Multimodal Model for Video and Robotics(12 posts)→
More from Multimodal
- Redditor struggles with outfit swap: Qwen Image 2.1 breaks body proportions — Existing-Box-7866 · 2026-10-27
- Moss Image Turbo 3-Step LoRA Tested: Fast Generation at 6-8 Steps — JadedScorpion · 2026-10-27
- Magnific teases October 28 release: 'We're making it move' — cuenca · 2026-10-27
- Midjourney --sref demo: retro Americana Batman art series — gizakdag · 2026-10-27
- AI video is a new medium, not cinema or animation — and it's perfect for depicting dreams — Kyrannio · 2026-10-27
- Dad turns his toddler's 31-year-old quote into a song via SUNO kids' choir — NathanpmYoung · 2026-10-27