NotchLive meeting model · In development

Built with public meeting records. Never with yours.

NotchLive is building a meeting-specific model on IBM Granite. It will run entirely on your Mac and work as an optional alternative to Apple Intelligence.

IBM Granite · On-device on Mac · No customer training data

NotchLive settings showing Apple Intelligence available now and the fully on-device NotchLive Local provider in development
Meeting-specific intelligence without cloud inference The model will process meeting text entirely on the Mac.
What NotchLive adds

A general model becomes a meeting model.

IBM Granite provides the foundation. NotchLive adds meeting-specific post-training, a structured meeting workflow, and fully local processing on the Mac.

IBM Granite foundation

IBM Granite provides the open foundation for NotchLive's meeting-specific post-training.

Meeting-specific post-training

NotchLive adapts Granite with curated historical public meeting records and the structured outcomes people need from meetings.

On your Mac

Meeting text will be processed locally. There is no NotchLive cloud inference service for this model.

Available today and coming next

Meeting Intelligence is already here. Another local option is coming.

Meeting Intelligence works today with Apple Intelligence on supported Macs. The NotchLive model will add another on-device choice rather than replacing the current experience.

The NotchLive model is still in development and is not included in the current release.

Training and evaluation data

Public meeting records in. Customer data out.

NotchLive's post-training and evaluation data comes from curated historical public meeting records that are public-domain or licensed for this use. It is separate from the IBM Granite base model's documented pretraining history.

What NotchLive can use

  • SourceHistorical meeting records released by public institutions or other documented sources.
  • RightsPublic-domain material or material licensed for post-training and evaluation.
  • ReviewProvenance, rights, format, and meeting relevance are checked before inclusion.

What NotchLive never uses

  • MeetingsCustomer audio, captions, transcripts, participant names, corrections, briefs, and saved meetings.
  • Personal workVoice Notes, Saved Notes, exports, imported libraries, and MCP activity.
  • Product signalsTelemetry and support conversations are never used for training or evaluation.
In-development NotchLive Local Meeting Intelligence view showing a public records review with decisions about provenance, licensing, and excluding customer data
A local model with a documented data boundaryPublic meeting records can shape the model. Customer meetings only produce your local brief.
Local processing

Your meeting is processed, not collected.

When you generate a brief, the selected model processes the meeting text locally on your Mac. The meeting stays in your NotchLive library and does not become training or evaluation data.

Input stays local

The meeting transcript is provided only to the selected local model on your Mac.

Output stays yours

The generated brief remains part of your local NotchLive meeting record.

Never becomes training data

Your meeting, edits, and generated brief never become training or evaluation examples.

Meeting-specific behavior

Built for meetings, not generic chat.

NotchLive is adapting Granite to follow a changing conversation, separate evidence from interpretation, and produce compact results people can review against the transcript.

Focus and key points

Track what the meeting is about and what materially changed.

Questions and actions

Separate open questions from commitments and next steps.

Entities and decisions

Keep people, organizations, topics, and decisions structured.

Language-aware summaries

Produce concise meeting output in the language selected for analysis.
Measured before release

We will publish what it can actually do.

Before release, NotchLive will test grounding, names, multilingual text, long meetings, action extraction, and visible uncertainty. Supported Macs and measured performance will be published with the model.

Grounded results

Generated claims should be traceable to the meeting record instead of introducing unsupported detail.

Language fidelity

Chinese scripts, CJK text, diacritics, names, and requested output languages should remain intact.

Clear limits

When evidence is weak, the model should preserve useful work and show uncertainty instead of filling gaps.

About the foundation model

IBM Granite underneath. NotchLive meeting intelligence on top.

IBM publishes Granite as an open model family under the Apache 2.0 license. Granite has its own documented pretraining history; the policy on this page covers NotchLive's additional meeting-specific post-training and evaluation data.

NotchLive meeting model FAQ

Is the NotchLive meeting model available now?

No. The NotchLive meeting model is in development. Meeting Intelligence available today uses Apple Intelligence on supported Macs.

Will the NotchLive meeting model run in the cloud?

No. The model will run entirely on the Mac. NotchLive will not operate a cloud inference service for this model.

Does NotchLive train on customer meetings?

No. NotchLive never uses customer meetings, personal notes, corrections, exports, MCP activity, support conversations, or telemetry to train or evaluate the model.

What data is used for meeting-specific post-training?

NotchLive uses curated historical public meeting records that are public-domain or licensed for post-training and evaluation. Source provenance and rights are reviewed before inclusion.

Will the model replace Apple Intelligence?

No. It will be a selectable alternative alongside Apple Intelligence, including for people whose Macs do not support Apple Intelligence and people who prefer another local model.

Meeting Intelligence is already here.

Use it today with Apple Intelligence on supported Macs. The NotchLive model is the next fully on-device option.