Product comparison
Local intelligence, supervised work
OCADE combines a local-first inference route with projects, files, current-information retrieval and permissioned computer work. This table compares product categories, not unverifiable competitor claims.
Official information
Read the details
Positioning
Choose OCADE when you want one workspace that can answer locally and prepare or execute supervised work. Choose a cloud AI workspace when remote hosted models are acceptable. Choose a standalone local runner when model execution matters more than an integrated workspace.
Feature comparison table
| Capability | OCADE | Typical cloud AI workspace | Standalone local runner |
|---|---|---|---|
| Default inference route | On-device desktop or browser engine | Remote provider | On-device model |
| Current information | Permissioned retrieval with source labels | Usually built in | Usually manual |
| Computer work | Permissioned, supervised, with STOP and receipts | Varies by service | Not normally included |
| Private synchronization | Optional; Device Only remains available | Account cloud by default | Usually manual |
| Offline behavior | Available where the selected local engine is ready | Generally unavailable | Core capability |
| Pre-launch price | $4.28 / $28 / $128 planned; checkout locked | Provider-specific | Software and hardware vary |
Key differentiators
- Local execution is the default route, not a privacy-mode add-on.
- Computer actions require explicit scope and keep STOP, PAUSE and TAKE OVER visible.
- Failure states name the unavailable engine or provider instead of silently switching to public-cloud inference.