Superwhisper is AI dictation: hold a hotkey, speak, and a local Whisper model (or a cloud model on your own keys) drops shaped text into the active Mac app. OpenCues is inline AI for typed text: a _ summons generation, rewrites and live data in place.
Where they overlap
Both respect the same instincts: local models, your own API keys, per-context customisation, and working inside every app rather than a chat window. Superwhisper's custom modes shape dictated output much like OpenCues cues and blanks shape typed text.
Where Superwhisper struggles
Insertion is a clipboard paste into the focused field, so its output is a one-shot drop: it shapes what you dictate but cannot rewrite text already sitting in the document, and nothing about the result is revertable beyond the app's undo. The mode system is powerful but hand-configured per context, and the experience is strongest on the Mac it was built for.
How OpenCues differs
Superwhisper is proprietary and macOS-centric; OpenCues is Apache-2.0 and runs across terminals, coding agents and Chrome. Superwhisper transcribes and polishes speech; OpenCues generates, transforms and computes on text you type. No competition for the microphone: OpenCues has no dictation, so the two coexist happily.
Both route models by context: Superwhisper per mode, OpenCues per feature. Superwhisper reaches any Mac app, while OpenCues covers terminals, coding agents and Chrome.
OpenCues is also extensible by design: cues and blanks are markdown files you write yourself, and blanks can bind shell scripts or runtime classes, all under an open standard anyone can implement. See write your own cue and write your own blank.
Related: Full comparison · Ollama vs cloud · Superwhisper