The Faloween contest is live. Make a monster on fal, and yours could replace the “?” in our Halloween trailer. You’ve got 48 hours to enter: 1. Create an original monster using any image model on fal, or ask fal Agent to make one for you. 2. Post it on Instagram or X, tag @fal, and add #faloween#. 3. Get your entry in before [Wed, Oct 28 at 11:59 PM PT]. The winner gets $500 in fal credits, and their monster stars in the full trailer, out October 31. A few rules: original monsters only, so no real people, celebrities, or characters from movies, games, or shows. The fal team will pick one winner. By entering, you agree to let us feature your monster in the film and our posts about it.

We've teamed up with @ComfyUI on their open-source app challenge: Build an open-source workflow on Comfy with FLUX 3 at its core to compete for our FLUX prize ($1K / $500 / $250 in API credits), part of $25K in total prizes. Entries close Fri 10/30.

FLUX 3 Max scores 1,139 Elo on third-party benchmark evaluations, placing it in the #1# tier of frontier image-to-video models based on independent human preference evaluations. fal Research is continuing to post-train the model, with further improvements to quality and performance on the way.

Meet the fastest model in the West. Introducing FLUX 3 Max, the latest frontier video model post-trained by fal Research in partnership with @bfl_ai. Tied for #1# in image-to-video quality. Among the fastest frontier video models. Now on fal.

Introducing FLUX 3 Max, the latest post-trained video model from fal Research in partnership with @bfl_ai. FLUX 3 Max is tied for #1# in image-to-video quality on independent third-party benchmarks, generating a 10-second, 720/768p video in ~12 seconds. For the next three weeks, FLUX 3 Max is 75% off, making it the most affordable frontier video model at its quality level.

FLUX 3 Max is 75% off for the next 2 weeks, making it the highest-quality, fastest and cheapest model for overall video generation tasks. Try it: Text to video: Image to video: Keyframes: Talk to us about building on, customizing, or licensing FLUX 3 →

Introducing FLUX 3 Max. Fastest in the west. Fast enough to try anything. And everything. Generate a 10-second video in 12 seconds. Tied for #1# in image-to-video. 22× faster than Seedance 2.5 at the same quality. FLUX 3 is a model designed for customization. @fal adapted and optimized it to create FLUX 3 Max.

Can’t stop, won’t stop. The open models keep dropping FLUX 3 Image Klein 5.6B was trained on NVIDIA GB200 NVL72 and can run on an RTX GPU. Big 🙌 to the @bfl_ai team. There's also an NVFP4 variant:

FLUX 3 Image Klein is here! Bounding-Box control, pixel-perfect editing, multi-ref and a lot more - with only 5.6B parameters. Weights are available on HF! This model is a massive leap forward for applications that need to run locally. Go use it 🌲

FLUX 3 Image Klein by @bfl_ai is now on fal Draw a box and get exactly that layout. Pick objects from any reference and place them where you want, edit with boxes, transfer styles and render text. Text-to-image and edit, Klein and Klein Base 👇

FLUX.3 Image Klein 6B just dropped on @huggingface 🔥 Base and distilled (16 steps), multi-reference editing, JSON bounding-box layouts play with it 👇

FLUX 3 Image Klein is here. Open Weights with the quality of a large frontier model. Customizable and fast. 5.6B parameters and runs on a single consumer GPU. Native 4K, bounding boxes, and pixel-perfect editing.

Introducing FLUX 3 Image. Control every pixel. Make precise multi-turn edits without changing any other pixel. Lay out the image exactly how you want using bounding boxes. Generate in up to 4K to preserve details. Use up to 10 references to compose an image. Commercial Weights available for companies running image generation at scale. Open Weights version of FLUX 3 Image is launching in the coming weeks.

Introducing FLUX 3 Action. An open weights 7B World Action Model that achieves first place on the RoboLab benchmark. It outperforms the previous best open model by 6.1 percentage points while using 56% fewer parameters and running up to 3.95x faster.⁠⁠ FLUX 3 Action removes the usual trade-off between world action model performance and VLA speed: it still predicts video and actions together, but plans more than twice as far ahead and runs faster per second of robot motion than the strongest open VLA. Teams can fine-tune FLUX 3 Action on their own demonstrations to create policies for a particular robot and task. Together with @nvidia, we also integrated FLUX 3 Action natively into @huggingface's LeRobot, with fine-tuning recipes included and edge deployment on NVIDIA Jetson. Beyond robotics, we’re also seeing promising results training task-specific policies for acting in simulated environments like gaming, controlling a vehicle, computer use, and wherever else a model needs to understand a visual environment and then choose what to do next. FLUX 3 Action builds on the same image, video, and audio pretraining as FLUX 3, but uses a smaller architecture designed for practical deployment. In midtraining, we trained the model to predict actions and future frames together. We’re releasing the weights, code, fine-tuning recipe, benchmarks, and reproducible examples so researchers and developers can build on the model with their own robots, environments, and tasks (see below).

FLUX 3 Video now has fast, precise editing. The fastest and lowest cost video editing model in the world. In our internal benchmarking, it’s also ahead of other leading models in the accuracy and precision of edits. Edit an existing video by: • Adding, removing, or replacing objects and characters • Replacing backgrounds • Editing text • Changing colors, materials, visual effects • Changing or translating dialogue with accurate lip sync • Changing the events of a clip