Black Forest Labs Launches FLUX 3 Video Model in Limited Release
Black Forest Labs launched FLUX 3, a multimodal AI model capable of generating images and video with audio up to 20 seconds from a single prompt, along with robotic vision and action capabilities. The model is entering limited early access with pricing and full benchmarks still unannounced, and open-weight versions will arrive later this year. Early preference testing shows FLUX 3 outperforming competitors like Luma Ray 3.2 and Runway Gen-4.5, though those results are labeled preliminary.
TL;DR
- FLUX 3 is jointly trained across image, video, audio and robotic action modalities rather than combining separate models
- Available through four product lines: FLUX 3 Video, Image, Action, and upcoming open-source FLUX 3 Dev
- FLUX 3 Video and Action entering gated early access now; Image rolling out in coming weeks; general availability timeline unspecified
- Preliminary benchmarks show 93% preference over Luma Ray 3.2 and 77% over Runway Gen-4.5 on 10-second video clips, but pricing and full methodology withheld until broader release
Why It Matters
FLUX 3 represents the first major multimodal model from BFL combining video, audio and image generation in a single architecture rather than stitching separate models together. The limited rollout and withheld benchmarking details mirror recent release strategies from Anthropic and OpenAI, but the delayed open-weight release breaks from FLUX's previous adoption pattern and may slow enterprise adoption.
Business Impact
Enterprises cannot yet calculate total cost of ownership or independently verify performance claims, as pricing and full benchmarks remain unpublished. The gated early access model creates friction for potential customers evaluating the technology against established competitors like Runway and Luma.
Key Implications
- Open-weight versions arriving later this year rather than alongside launch may disadvantage developer adoption compared to previous FLUX releases
- Preliminary benchmark results describe a pre-release checkpoint, not the shipping model, creating uncertainty about actual customer-facing performance
- Lack of published evaluation methodology and sample sizes prevents independent reproduction of video comparisons, limiting enterprise confidence in claims
What to Watch
Monitor the timing and scope of FLUX 3 Image general availability and the feature set of FLUX 3 Dev when it launches. Track whether BFL publishes full benchmarks and pricing before or after broader rollout, and observe adoption velocity among enterprises once early access expands. Watch for competitive responses from Runway, Luma, and Google on multimodal video generation.
Subscribe to the newsletter
The latest stories and analysis, delivered to your inbox.
Free. No spam. Unsubscribe any time.
