AIToday
Large Language ModelsOpen-Source AITechCrunch AIPublished: Sep 25, 2026, 13:00 JST

PrismML's 1-bit Bonsai LLM lands on Snapdragon AR1 glasses

PrismML's 1-bit Bonsai LLM lands on Snapdragon AR1 glasses

3 Key Points

  1. What happened

    PrismML created a version of its tiny language models for smart glasses running on Qualcomm's Snapdragon chips. At Qualcomm's Snapdragon Summit, the chipmaker showcased PrismML's 1-bit Bonsai LLM, a 2-billion-parameter model tuned for vision and language that runs locally on AI smart glasses built on the Snapdragon AR1 Gen 1 Platform.

  2. Why it matters

    Running the model locally on the device, rather than in the cloud, is the point of this showcase — it signals that PrismML's open-weight, on-device approach is being paired with Qualcomm's smart glasses hardware.

  3. What to watch

    No smart glasses running PrismML have been announced yet, so the showcase has not yet turned into a shipping product. Watch whether a device maker commits to the Snapdragon AR1 Gen 1 Platform version.

WHO IT HITSSmart glasses makers and AR hardware teams evaluating Qualcomm's Snapdragon AR1 Gen 1 Platform are the ones most directly affected, since PrismML's on-device model is now showcased for that chip, though no shipping glasses have been announced.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, in your inbox every morning.

Context & Analysis

PrismML was founded by Caltech researchers and is advised by UC Berkeley's Ion Stoica. Its claim to fame, as TechCrunch previously reported, is shrinking larger models substantially — in this case by 4x — while retaining almost all of their performance on standard benchmarks. That background explains why the company's pitch centers on open-weight AI that runs on devices rather than in the cloud.

The Qualcomm showcase is a step toward that vision: pairing a small, locally run model with a chip platform built for smart glasses. The startup frames this as an alternative to relying on the privacy promises of proprietary AI labs and their need for more compute.

The stakes here hinge on whether the Snapdragon AR1 Gen 1 Platform version moves from a summit demo to an actual device. No smart glasses running PrismML have been announced yet, so it remains to be seen whether a hardware partner picks it up — that would be the point at which the on-device pitch reaches real wearers rather than a stage.

FAQ
What kind of model is PrismML's 1-bit Bonsai LLM?
It is a 2-billion-parameter model tuned for vision and language, designed to run locally on AI smart glasses so wearers can ask what they're looking at in real time.
Can I buy smart glasses running PrismML now?
No. The model was showcased by Qualcomm at its Snapdragon Summit, but no smart glasses running PrismML have been announced yet.
What is PrismML's larger goal?
The startup is pursuing open-weight AI that runs on devices and makes better use of the computing power they already have, pitched as an alternative to depending on proprietary AI labs' privacy promises and their need for more compute.

Get the latest Large Language Models news every morning

For example, today's edition would include:

  • Meta's Muse Tops Apple's App Store; Bank, Insurer Stocks TumbleYahoo Finance AI · 1h ago
  • Schmidhuber joins Sakana AI to lead RSI LabITmedia AI+ · 4h ago
  • TypeSafe AI launches Jev, claiming up to 200x faster inference without hallucinationsITmedia AI+ · 4h ago

AI-summarized, only the topics you pick: one digest a day via Email, LINE, or Slack.

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.

Questions and answers are published on this page.

Related Articles

Next articleDataiku launches Agent Management for cross-vendor AI agents