PrismML launches 1-bit Bonsai LLM for Qualcomm smart glasses
PrismML, a lab founded by Caltech researchers and advised by UC Berkeley’s Ion Stoica, unveiled a version of its tiny language models that can run on Qualcomm‑powered smart glasses. At Qualcomm’s Snapdragon Summit on Wednesday, the company demonstrated its 1-bit Bonsai LLM operating locally on the Snapdragon AR1 Gen 1 Platform. The smart‑glasses variant is a 2-billion-parameter model tuned for…
Key points
- PrismML demonstrated its 1-bit Bonsai LLM on Qualcomm’s Snapdragon AR1 Gen 1 Platform at the Snapdragon Summit
- The smart‑glasses version is a 2-billion-parameter model tuned for vision and language
- PrismML claims the model is four times smaller than larger equivalents while retaining near‑full benchmark performance
PrismML says the model shrinks larger models substantially – in this case, by 4x – while keeping almost all performance on standard benchmarks. The startup frames the release as an open‑weight alternative to relying on proprietary AI labs that demand ever more compute. No smart glasses that ship with PrismML’s model have been announced yet, so the claim remains untested in a commercial product.
Model page: 1-bit Bonsai LLM →
The story so far
2 episodes →- PrismML launches 1-bit Bonsai LLM for Qualcomm smart glassesthis story
PrismML brings its tiny LLMs to Qualcomm-powered smart glasses
TechCrunch AI · 24 September 2026
Loading the full article…
This text was published by TechCrunch AI and written by Julie Bort. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Generative AI & Models
All →- Anthropic releases Claude Opus 5.5; OpenAI halves prices for GPT-6 Sol and Luna · 14 src
- OpenAI releases ChatGPT Images 2.5 with sketch, point, and template tools · 4 src
- FreedomIntelligence releases HuatuoGPT-3 multimodal model on Hugging Face · 1 src
- BottleCap AI cuts Qwen3.8-27B’s reasoning tokens by 37.2% with ThinkingCap-Qwen3.8-27B · 1 src
- What has changed with GPT-6 Astra? Basic specifications of the new model that supports 'thinking, researching, and creating' in ChatGPT · 2 src
Comments
via GitHub Discussions