Two free editors that run in your web browser — one for video, one for photographs — each with an AI agent that can do the editing for you, using your own OpenAI or Gemini key.
Free during beta · nothing to install · one key covers both
Two editors, one studio
A multi-track timeline. Import footage, trim and split, apply effects, transitions and colour correction, track motion, add titles and captions, and export MP4 or WebM — all rendered in the browser.
Layers, selections and pixels. Paint and retouch, cut things out, correct colour non-destructively, add text and shapes with layer styles, and export PNG, JPEG or WebP — again, entirely in the browser.
They share an account, an AI configuration and a design. Add your API key once and both editors use it; sign in once and both are unlocked.
Goblin Studio is a pair of media editing applications. In the video editor you import video, audio and image files, arrange them on a multi-track timeline, trim and split clips, apply effects, transitions and colour correction, add titles and captions, and export the finished result as an MP4 file. In the photo editor you open an image, work on it in layers, select and mask parts of it, paint and retouch, correct colour and tone, add text and shapes, and export a PNG, JPEG or WebP. All editing, preview and rendering happen locally in your browser — your files are never uploaded to our servers.
What makes them different is the built-in AI agent. Instead of performing every edit by hand, you can describe what you want in plain language and the agent carries it out using the same edit operations available to you — so anything it does, you can adjust or undo. It runs on an OpenAI or Google Gemini API key that you supply and control, and one key serves both editors.
Signing in with a Google account is required for two features only: exporting your work and using the AI agent. Connecting Google Drive is entirely optional, and is used so you can import footage you select from your Drive and save your project files there to continue an edit on another computer. Everything else works without an account.
The agent
Describe the change. The agent performs the same edit operations you would — every one typed, undoable, and recorded against its name.
Bring your own AI
No bundled model, no AI subscription, no credits to buy. Point it at OpenAI or Google Gemini with your own API key and pick the model for every job.
Switch provider whenever you like.
Agent, vision, image, speech — chosen separately.
You pay the provider. We never touch it.
Your key is stored server-side, never sent back to the browser.
A real editor underneath
Trim, split, ripple-delete, group, nest.
Keyframable, with chroma key and colour work.
Move, scale, rotate, crop in the preview.
Styled text, auto-captioned from the audio.
Turn written lines into spoken audio.
Rendered in the browser with WebCodecs.
Your media and projects live in your own browser. Editing, preview and export all happen on your machine — we never receive your footage.
Optional
Import footage you pick, and save projects so you can pick an edit back up on another machine.
drive.file — the narrowest scope there is. It cannot browse or
read anything else in your Drive
Free during beta