Black Forest Labs Launches FLUX 3: Unified Multimodal Model for Image, Video, and Audio
bfl_ai · x · 2026-08-05
Black Forest Labs has officially released the FLUX 3 preview, a multimodal model trained uniformly across image, video, and audio. Its core highlight is the ability to generate video with synchronized audio via a single request shape (endpoint).
Core Features & Specs
- Synchronized Audio: Generates multilingual speech with strong lipsync, alongside sound effects and ambience directly with the frames.
- HD Long Video: Capable of generating up to 20 seconds of FHD (1920×1088) video at 24fps in a single request.
- Multi-Scene & Typography: Supports multiple scenes and camera angles in one generation, with accurate in-scene text rendering. Stylistically versatile beyond cinematic looks (e.g., animation, motion design).
Supported Modes
All requests run on the same flux-3-video endpoint, differentiated by a mode parameter:
- Text to Video (t2v): Generates a clip from a text prompt.
- Image to Video (i2v): Uses input images as frames to build the clip.
- Video Continuation (v2v): Extends an existing video clip forward.
The official blog notes that video editing and Omni Reference with images and videos will be available soon.
More from Multimodal
- Kandinsky Lab open-sources S2V 1.0: audio-driven avatar video from one still, MIT licensed — lmoroney · 2026-10-16
- Meme: Flux 3 Klein 5.6B steals the spotlight from Flux 2 Klein 9B — xb1n0ry · 2026-10-16
- Pichai amplifies user demo: Gemini Argon builds Asuka-period pagoda in 11 hours — sundarpichai · 2026-10-16
- Midjourney style code --sref 1127756965 yields bold collage-style illustrations — sergeantsref · 2026-10-16
- A cinematic ocean-to-cliff drone-style prompt worth trying on video models — umesh_ai · 2026-10-16
- Flux.3 klein vs Qwen Image 2.1 turbo on RTX 4070: 10s vs 3s per 1MP edit — Ant_6431 · 2026-10-16