Introducing Diffusion Studio Pro. The world’s first multimodal AI studio for professionals. Think CapCut + Cursor. Local first, browser based editor (non linear). Free unlimited tier. Fully agentic sidebar with access to 16+ generative models. Comment “DSPRO” to get early access!

We're rolling out our waitlist right now. Sign up at and join the new era of filmmaking.

About the project: We started Diffusion Studio in March 2023 with the mission of revolutionizing video editing through text-to-video. Due to the lack of suitable video processing frameworks, we spent most of the time building an AI-native video editing engine from scratch, optimized for agents and LLMs. Clients like Topaz Labs are already using it in production. After a significant breakthrough in agentic video processing a couple of weeks ago, we decided to go public with our consumer-facing web app.

We’re solving 2 big pain-points: - Finding or recording editing assets is time-consuming. Many people don’t want to be on camera, and often have no other way to share their voice publicly. With DS Pro, you get access to all the generative models needed for filmmaking in a single chat: image generation, image-to-video, lip sync, text-to-music, text-to-SFX, and more, all easily accessible. ⁃ Video editing is hard and time-intensive. With DS Pro, you can edit your assets through chat. The video editing agent learns from your editing style and supports your workflow, keeping you in the director’s seat.

What we’re adding before the public beta: - Agentic access to the media pool, allowing the agent to visually read video, images, and audio. This will enable automatic footage processing and rough cut generation. - Visual composition feedback, enabling the agent to visually analyze the editing state and create a better feedback loop for its decisions. - Next timeline action prediction - when you place the cursor on the timeline, DS Pro will suggest the most likely next editing action.

A huge thanks to @FAL and @replicate for the grants, which significantly reduced the time it took to integrate state-of-the-art video generation models and allowed us to test the agent more effectively.

@replicate Tagging a few folks who might find this useful: @rileybrown_ai @PJaccetturo @EHuanglu @javilopen @WuxiaRocks @Ror_Fly

Discover the top Safari web extensions to boost productivity, enhance workflow, and customize your browsing experience on Mac, iOS, and iPadOS.

Congrats on launch! I'm a big supporter of streamlining and automating video editing through writing code (and in-browser video editing). I tried the diffusion studio library 4-5 months ago and was amazing by its code quality. Wishing you and the team the best of luck!

DSPRO

DSPRO! congratulations, I am so proud of the team!

🙏🙏🙏