We are thrilled to announce the beta release of , bringing the most advanced Stable Diffusion model into a highly controllable real-time video workflow.

Better yet: and available for the real-time AI video community to explore and extend.

Builders and Creative Technologists can now generate directly on the Daydream platform, with the potential to achieve up to 25 FPS in optimized configurations.

dotsimulate and others are already creating SDXL-based tools and applications using the Daydream API or self-hosting our open-source StreamDiffusion fork.

Our open source stack allows us to rapidly combine many parallel research tracks and deliver groundbreaking quality and controllability.

Here a few of the key components - already battle-tested by Daydream builders:

let you use any reference image to guide your video's style - similar to LoRAs but with real-time control.

We’ve accelerated HED, Depth, Pose, Tile, Canny ControlNets for incredibly granular control of the workflow’s spatial awareness. You can use multiple at the same time, and adjust the strength of each.

SD1.5 is a community favorite for a reason - and we’ve coupled it with accelerated IPAdapters for a high-framerate style transfer experience.

Daydream is a growing community of researchers, builders and artists, steering the future of open-source real-time AI video and world models like StreamDiffusion and MatrixGameV2.

Much of across the reaches of the internet, and to help builders, artists, and researchers accelerate their work.

We've compiled some resources to start, but open source is most powerful when a community collaborates;

The Daydream community has created some amazing experiences and visuals with the Playground - but it's been a little too difficult to get good results.

Starting from scratch with all the knobs can be overwhelming and error-prone, so we’ve made it simpler and more fun. Now you can:

You don’t need to know every parameter to get great results. Now, you can build on community creations, tweak them, and share your own experiences.

To help you get started, we’ve curated a few amazing configurations created by Daydream community members. You can try them here.

No more guesswork for getting good results: now you can by testing out some community creations, , and for others to build on.

Share a look you love, try someone else’s setup, or collaborate to push an idea further.

Creating detailed prompts isn’t always easy- that’s why we added image-based controls.

With IPAdapters (Image-Prompt Adapters), you can just shuffle images or upload your own reference, then adjust intensity to instantly change the style or vibe. IPAdapters behave a bit like LoRAs, but live and versatile.

Now with and plus faster , you can guide your video’s look in real time using any image.

By combining and , you get two simple but powerful ways to unlock creativity: start from a proven recipe, then swap in your own reference images to instantly make it yours.

Added strict parameter validation to prevent malformed requests. Previously, the API silently accepted requests with incorrect parameters, incorrect types, or unknown parameters. Now, any of these validation failures will return a 400 Bad Request response with a descriptive error message indicating the specific issue.

Requests that were previously accepted despite having invalid parameters will now be rejected. Ensure all API calls use correct parameter names, types, and values.

We're removing the biggest friction point for new developers - waiting for Daydream to provision an API key. Now anyone can sign up and start building immediately.

This dashboard enables builders to generate, manage, and organize their API keys.

Existing builders can access their keys immediately; new signups get instant self-serve access.

Future improvements will include usage tracking, debugging tools, billing, and team management.