made this with text to lottie and claude code my workflow: design something in figma or just make a rough draft. then ask claude to animate it you'd be surprised how easy it is to add easings, masks, transforms, text effects and all the motion details you can also ask claude to export it as a gif, mp4, png sequence, resize it in different aspect ratios claude and codex already creates something with taste and if you know what you want to create, this workflow is really useful and btw, it's free

Introducing text-to-lottie: an open source skill and harness for generating production ready Lottie animations with codex/claude code. $ npx skills add diffusionstudio/lottie Prompts guide and repo in the comments.

looks like we found the one went with right some slight adjustments what do you think?

We must be the first browser-based editor that supports image sequence playback

this video made with Diffusion Studio adjusting the aspect ratio and tweaking the layout for it took less than 2 mins one project, multiple explorations, every aspect ratio on one infinite canvas never been this easy

Editing was overdue for a rethink. We built it the way design tools work @diffusionhq is now in Open Alpha An infinite canvas for designing, editing, exploring, generating, and exporting Runs in your browser and locally

video editing is stuck in 2005, time for something new introducing diffusion the first infinite canvas for video and motion graphics like figma, but for editing

deep in the details

prototype of anchor point controller experience for @diffusionhq

People kept asking under my viral tweet: if it's not Cursor for video editing, what exactly are we building? Short answer: the missing pieces. I.e. the VS Code for video editing. In Sep 2025 we hit reset. We gathered everything we'd learned and started a full redesign. That included migrating the front end from React to SolidJS. A tech deep dive is coming. The new redesign is on track to ship in the next two months. It needs to hit a few core goals: - Low floor, high ceiling pro NLE - Local first webapp & Electron native - Runs hardware accelerated/ GPU backed in chromium - Ships with most table-stake features required for professional editing - Native Media Gen integrations - Basic chat based editing Phase 2 (Q2 2026) will bring: - A highly extensible plugin API - Multiplayer - Color grading - Pro grade asset lib (sfx, b-roll etc.) - Viral WebGPU shaders: dither, ascii, liquid glass etc. Some editors don’t believe a real pro NLE can live in the web. We’re about to prove it can run faster and feel better than the native apps people rely on today.

Introducing Diffusion Studio Core v4. A fast, browser based video compositing engine powered by WebCodecs. We built it for our own pro non linear editor. Now you can use it as the foundation for your own timeline based app. GH below 👇

Just shipped v0.21.0 of Diffusion Studio Pro, featuring the new Captions Clip. You can cut and trim it like any video clip and apply different templates to each section. Here is a quick look at how it works 👇

Diffusion Studio Pro 0.16.0 introduces markers, which you can now add and select to define your export range. We've also added the option to render audio only, or video without audio.

It was a pleasure merging this bad boy.

Diffusion Studio Pro is no joke. Even if you take out all the AI integration (which is insane), you're still left with a very capable, modern, 100% client-side video editor web app. I know because I actually helped develop it! Don't sleep on it 👀

No login, no waitlist. Just instant video editing in your browser.

Just shipped our 135th timeline redesign

WebGPU inference might be the most underrated tech in AI right now. This entire audiobook was generated in the browser on my M1 MacBook Pro. - Kokoro 82M for TTS - Whisper Base for transcription/captions It runs so fast with WebGPU that most people won't even realize it's happening on their own hardware.

We added a feature to lip sync footage on the timeline in Beta 0.14.0