Edit by talking to AI
the AI video editor
A captioned video in as little as 5 minutes. Talk to AI on the timeline to edit — and let AI fine-tune your subtitles too. The conversational AI that drives editing runs on your existing Claude / ChatGPT — no extra fee (BYOK & local too). Generative AI for images and video is pay-as-you-go with credits. Public release is planned for July 2026 (free beta now).
Point. Click. Direct.
Talk to AI while you edit. Point at a clip, say what you want — AI takes it from there.
Click on the Timeline
Point at a clip and say 'make this shorter.' AI knows exactly where you're looking.
Select a Range
Drag across the timeline and say 'cut the dead air here.' Precise and instant.
Drop a Reference
Show a reference video and say 'match this style.' AI reads the visual language and applies it to your footage.
Missing footage? Generate it on the spot.
While you edit, generate the video, images, and sound you need with AI — dropped straight onto your timeline. Just describe it, and create it your way.
Generate video between frames
Set a clip's start and end frames, and AI generates the in-between as video. Drop missing cuts and transitions straight onto the timeline.
Generate and place images
Create backgrounds, thumbnails, and insert images by talking. Specify the style and place them directly on clips or as overlays.
Generate music, SFX & voice
Generate scene-fitting BGM, sound effects, and narration (TTS) with AI and place them. Not just visuals — sound, exactly how you want it.
See it in action.
Talk to the timeline, watch it move. The full experience, in a short video.
Demo video coming soon
Video editing should take hours, not days.
You lose time bouncing between tools, re-explaining yourself to AI, and waiting on cloud uploads — more logistics than actual editing.
Your whole day goes to switching tools
Generate, edit, export — each in a different app. A 4K upload alone takes minutes. Every context switch kills your focus.
AI doesn't remember your last 100 videos
You re-explain your style, audience, and tone every single time. No memory. Every prompt starts from zero.
Cloud means waiting and paying
10 GB upload, 20-minute queue, per-minute charges. Your raw footage shouldn't have to live on someone else's server.
The AI brain you already pay for.
Already on Claude Pro or ChatGPT Plus? Connect it as Akari's brain — no extra AI subscription required. Editing is always free. Pay only when you generate.
Connect Claude or ChatGPT (BYOS)
Already paying for Claude Pro or ChatGPT Plus? Connect it via a secure local bridge and that subscription becomes your editing brain. Zero additional charges from Akari.
Bring Your Own API Key (BYOK)
Plug in your OpenRouter, Anthropic, or Fal keys directly. No Akari markup. You pay the provider only for what you use — full cost control, your way.
Run Locally with Ollama
No subscription. No API key. No internet. Edit entirely offline with local models. Your footage and drafts never leave your Mac.
Your Data Stays on Your Device
Akari is edge-native by design. Your Mac is the center — not a fallback. Raw footage, client materials, and unreleased projects — none of it leaves your Mac without your permission.
- All memory and data stored locally
- No cloud sync unless you opt in
- AI queries through encrypted API only
- No conversation data stored on servers
- Full data portability — export anytime
- Works offline — no internet required for core features
Akari Video is the entry point. Akari OS is the depth.
Behind Akari Video runs a 7-agent team — Partner, Studio, Operator, Researcher, Guardian, Memorist, and Analyst — plus a shared knowledge store called Pool. Everything runs on your Mac.
Akari Video
The AI video editor built for creators. Just say "cut here" and Studio handles it. Connect your existing Claude or ChatGPT subscription as the brain — zero extra AI cost. Generative AI (images, video) runs on credits, only when you use it.
Explore Akari VideoComing Soon
Writer App
Articles, posts, long-form — inside the Shell
Pool Browser
Visual graph of your shared knowledge
Memory Viewer
See what your Memorist remembers (AMP)
Image App
Independent app for design & graphics
Social App
Independent app for multi-platform publishing
Be among the first creators
Akari Video is still in development. We're looking for the first users to help shape it — your feedback will define what it becomes.
First access before launch
Try Akari Video before the official release. Core editing tools and the conversational AI are free from day one of your beta access.
Your feedback shapes the product
Share feature requests, bug reports, and honest impressions directly with the dev team. A closeness only possible at pre-launch.
Starter credits for generative AI
Beta participants will receive a starter pack of credits for generative AI (image, video, audio). Everything else — no extra charge.
No credit card required. Editing and conversational AI features are free throughout the beta period.
Questions answered
Everything you want to know before joining the beta.
How does pricing work?
Video editing — cutting, timeline, export — is free. AI conversation (giving instructions, chatting) is also free when you bring your own Claude, ChatGPT, or similar subscription or API key. The only thing that costs credits is generative AI: AI image, video, and audio generation. These consume credits as you use them ($1 ≈ 100 cr).
Why is the conversational AI free?
Akari Video uses a Bring Your Own Key (BYOK) model. If you already subscribe to Claude Pro, ChatGPT Plus, OpenRouter, or similar, you simply connect your key and the conversational features are yours at no extra charge — Akari incurs no API cost, so we pass that along to you. Local LLM support (Ollama, etc.) is also planned.
How do credits work?
Credits are consumed only by generative AI features that require a model call — AI image generation, AI video generation, AI voice generation, etc. Pro includes 2,000 cr/month ($19), Max includes 8,000 cr/month ($59). New accounts receive 100 cr free (for transcription only). Additional credits can be purchased at any time.
What systems are supported?
Akari Video currently runs on macOS with Apple Silicon (M1 and later). Because video processing happens directly on your Mac's hardware, Intel Macs and Windows/Linux are not supported at this time. We'll announce expanded platform support when available.
What does "beta" mean, and when does the full version launch?
We're currently in a closed beta with a waitlist. The full product launch is planned for July 2026. Everyone who signs up during beta will receive priority early access. We're building this together — your feedback shapes what ships.
Where is my data stored? Is my privacy protected?
Your video files and edit data stay on your Mac. Nothing is uploaded to the cloud unless you explicitly use a generative AI feature, which temporarily sends only the data needed for that operation. Your data is never used for model training or shared with third parties.
How do cancellation and refunds work?
You can cancel anytime; access continues until the end of your current billing period. Top-up packs are valid for 180 days from purchase. Payments are generally non-refundable, except where required by law or for issues attributable to us (e.g., double charges). Payments are processed via Polar.
Can I use my own API key (BYOK) or a local LLM?
Yes. You can enter your Claude (Anthropic), OpenAI, or OpenRouter key in Settings to use your existing subscription at no extra cost. Local LLM integration (Ollama and others) is in development. Your API keys are stored securely in macOS Keychain and are never transmitted externally.
Start editing with AI today.
Video editing is free. Connect your existing Claude or ChatGPT subscription as the brain — no extra AI charges. Pay only for generative AI (images, video, audio) as you use it with credits. Get started on your Mac with the free Akari Video beta.
Public release planned for July 2026 (now in free beta) · No credit card required · macOS 12+ · Your data stays on your Mac