AI
[AINews] Black Forest Labs FLUX 3 - Multimodal Flow Models that beat Seedance 2.0, Gemini Omni and Grok Imagine
Black Forest Labs has officially launched FLUX 3 Video, a unified multimodal model featuring native audio generation across all outputs.
![[AINews] Black Forest Labs FLUX 3 - Multimodal Flow Models that beat Seedance 2.0, Gemini Omni and Grok Imagine](https://cdn.ravenclip.com/channels/79/575/images/0-mDVpY2Ir.jpg)
Key takeaways
- Black Forest Labs launched FLUX 3 Video with native audio generation across all modalities.
- FLUX 3 Video supports up to 3-minute one-pass generation and agentic chaining of clips.
- Moonshot AI's open-weight Kimi K3 is a 2.8T-parameter model that found 23 out of 26 vulnerabilities on Aikido Security's private benchmark.
- US frontier labs' safety guardrails and API-only access may reduce their usefulness for defensive vulnerability analysis compared to Chinese open-weight systems.
- OpenAI shipped desktop voice control for ChatGPT Work and Codex.
Black Forest Labs has officially launched FLUX 3 Video, a unified multimodal model featuring native audio generation across all outputs. The model supports text-to-video, image-to-video, video-to-video, and agentic chaining of clips into multi-shot sequences. Separately, Moonshot AI's open-weight Kimi K3 model, boasting 2.8 trillion parameters, has matched OpenAI's frontier models on a private cybersecurity benchmark, fueling discussions on the defensive utility of open-weight systems.
In their words
“Its core capabilities include the following (all outputs come with native audio generation):Text-to-video generation.”
By the numbers
- 3-minute
- One-pass generation capability of FLUX 3 Video
- 2.8T
- Parameter count of Moonshot AI's Kimi K3 model
- 23/26
- Vulnerabilities found by Kimi K3 on Aikido Security's benchmark
- $250
- Discussed payout figure in Authors Guild class action
How it unfolded
- Black Forest Labs launched Flux 1 hinting at video
- Black Forest Labs launched FLUX 3 Video
Turn stories like this into views
Ravenclip finds the AI news, makes the video, and posts it before attention moves on.
Common questions
- What happened with Black Forest Labs?
- Black Forest Labs launched FLUX 3 Video with native audio generation across all modalities.
- Where can I read the original report?
- Read the full report at latent_space.
