● Covered by 1 source · 2 reportsMedium impact2 neutral
PrismML Releases Bonsai 2 27B, a Compressed LLM for PCs and Smartphones
Key points
- PrismML released Bonsai 2 27B, a compressed LLM.
- Bonsai 2 27B is 5.9 GB, fitting on PCs and smartphones.
- It retains 98% of Qwen3.8 27B's benchmark performance.
- The model offers a 9x to 10x memory reduction.
- PrismML developed a Bonsai LLM version for smart glasses with Qualcomm Snapdragon chips.
- Qualcomm showcased PrismML's 1-bit Bonsai LLM at its Snapdragon Summit.
- The smart glasses model is a 2-billion-parameter model tuned for vision and language.
- The smart glasses model runs on the Snapdragon AR1 Gen 1 Platform.
PrismML Introduces Bonsai 2 27B
PrismML, an AI lab, released Bonsai 2 27B, its latest compressed large language model. This model is a compact version of Qwen3.8 27B, an open-source model from Alibaba.
Technical Specifications and Performance
Bonsai 2 27B is compressed to 5.9 GB, which allows it to run on personal computers and potentially high-end smartphones. This represents a 9x to 10x reduction in memory footprint compared to the original Qwen3.8 27B model. PrismML states that Bonsai 2 matches 98% of Qwen's aggregate benchmark scores, an improvement from the first Bonsai model which achieved 95%.
Company Background and Funding
PrismML was founded by Caltech researchers, with CEO Babak Hassibi, a Caltech professor and compression expert, leading the company. Ion Stoica, co-founder of Databricks, serves as an adviser. The company has raised a $22.25 million seed round and is backed by investors including Khosla Ventures, Cerberus Capital, and Caltech.
Implications for On-Device AI
The development of highly compressed LLMs that maintain performance could expand the possibilities for running AI models directly on user devices. This approach reduces reliance on cloud infrastructure and could enable more private and responsive AI applications on PCs and mobile devices.
Updates
- PrismML developed a Bonsai LLM version for smart glasses with Qualcomm Snapdragon chips.
- Qualcomm showcased PrismML's 1-bit Bonsai LLM at its Snapdragon Summit.
- The smart glasses model is a 2-billion-parameter model tuned for vision and language.
- The smart glasses model runs on the Snapdragon AR1 Gen 1 Platform.
✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →
The daily brief
One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.
One email a day. Unsubscribe in one click, any time.
Today's brief
Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.
▶ Play today's brief
New every morning, and the back catalogue is archived by date.
How outlets covered it
PrismML, an AI lab, has developed a version of its compact language models for smart glasses utilizing Qualcomm's Snapdragon chips. This development allows for local execution of AI models on devices, reducing reliance on cloud-based processing and addressing privacy concerns.
PrismML released Bonsai 2 27B, a large language model compressed to 5.9 GB, making it suitable for PCs and high-end smartphones. This model achieves 98% of the benchmark performance of the original Qwen3.8 27B model, representing a 9x to 10x memory reduction. The development of highly compressed yet performant LLMs could enable broader on-device AI applications.