Topics · 6 stories

Local AI

Every 7am AI story on this topic, newest first. Each links to its sources, labelled by how sure we are.

  1. llama.cpp serves several users at once almost three times faster

    Running several agents on one local model just got much cheaper in waiting time.

  2. A free update lets big local models read prompts much faster

    The wait before the first word is what you notice with long documents, and that part shrank.

  3. A small open model turns scanned documents into clean text on your own machine

    Private paperwork can be made searchable without sending it to a cloud service.

  4. NVIDIA’s free model transcribes 40 languages, and llama.cpp learns to talk

    Speech can now be turned into text without the audio ever leaving your office.

  5. Liquid AI’s small LFM3 models hear, see and speak on a laptop

    A voice assistant can work offline, so what you say never leaves the room.

  6. Baidu’s free OCR model reads a whole book in one go

    Long contracts or records can be read on your own hardware and never leave the building.