One meeting-tuned model, on your device — unlimited.
Zolash includes a meeting-tuned AI model that runs entirely on your device. There's no cloud inference, no API key to bring, and no account to configure — and because the compute is yours, there are no AI credits, no per-minute charges, and no per-meeting caps. Record as much as you like; the price never moves.
Available for macOS, Windows, and Linux. Nothing uploaded.
One meeting-tuned model
A model built for meetings — not a general chatbot bolted on.
Zolash ships with a model tuned for the work of meetings: transcribing speech, labelling speakers, and drafting summaries and action items. It's the same model for everyone, it lives in the app, and it runs on your machine — no API key to paste, no account to wire up.
Unlimited by design
Record all day. The bill never moves.
Because the model runs on your hardware, there are no AI credits, no per-minute charges, and no per-meeting caps. A three-hour workshop costs exactly the same as a ten-minute standup: nothing extra. Heavy users are never penalized for using the product.
Why local matters here
No cloud inference means nothing is sent, and nothing is metered.
Running the AI in the cloud is what forces other tools to upload your transcript and meter your usage. Zolash runs the model on your machine, so producing a summary sends nothing off the device — and there's no cloud compute to bill, which is exactly why it can be unlimited.
See the privacy architectureWhat’s included
- Meeting-tuned model included — nothing to configure
- Runs on your device: no cloud inference, no API keys
- Unlimited meetings — no AI credits or per-minute metering
- Same model for every user; your data never tunes it
- No account required for the AI to work
- Works offline, like the rest of the app
Questions, answered.
Do I need to bring my own API key or model?
No. The meeting-tuned model is built into Zolash and runs on your device. There are no keys to paste and no external model to connect — it works out of the box, offline.
Are there usage limits or AI credits?
No. Because the model runs on your own hardware, there are no AI credits, no per-minute charges, and no per-meeting caps. Extra hours of meetings cost nothing.
Does my data improve the model?
No. The model is the same for every user and is never trained on your meetings. Your conversations stay on your device and are never sent anywhere to tune it.
Is a local model powerful enough for good summaries?
Yes. The model is purpose-built for meetings and runs at full quality on your machine. It's specialised for exactly this job rather than being a general-purpose chatbot, which is what lets it run well locally.
Unlimited meetings. On-device. No credits.
Start your free trial. The on-device model is built in — nothing to configure, nothing to upload.
Available for macOS, Windows, and Linux. Nothing uploaded.