Edit without uploading footage
Import media, organize assets, build sequences, make timeline changes, review the result, and export from the native Mac app. AVE does not require a cloud project before you can start editing.
AVE starts from the Mac: local project folders, local media, local analysis, timeline editing, and native export. Work offline with local tools, then connect an external provider only when the job calls for it.
AVE does not use “offline” as a blanket promise. The core editor is local; optional external intelligence is clearly separated.
Import media, organize assets, build sequences, make timeline changes, review the result, and export from the native Mac app. AVE does not require a cloud project before you can start editing.
Transcription, visual analysis, footage retrieval, and project context can stay in the local workflow. Installed local models can provide AI planning without sending the project to an AVE service.
Claude, Codex, custom API endpoints, and image or video generation providers are optional connections. Those services need a network and operate under their own terms.
Project files, imported media, local analysis, timeline operations, and exports stay on your Mac by default.
External assistants, custom API endpoints, and BYO generation providers are optional choices.
Local models and installed assistants can be used for planning when configured.
Provider keys are handled by the desktop app and stored using encrypted secure storage where supported.
The AVE 1.2.3 Mac download is distributed through GitHub Releases.
Local-first does not mean pretending the network does not exist. It means the local editor is the default, while external services are explicit connections. Read the full privacy policy and security details before connecting a provider.
Project files, imported media, transcripts, visual analysis, Footage Memory, timeline changes, renders, and exports remain in AVE's local workflow. Importing footage does not silently upload it to an AVE cloud workspace.
If you connect an external assistant, custom API endpoint, or generation provider, the selected service may receive the prompts and context needed for that request. You choose the connection, credentials, and provider terms before using it.