# PrismML releases 5.9 GB Bonsai 2 model compressed from Qwen 27B

_Published Thursday, September 17, 2026 at 8:10 PM EDT · AI, Capital · Latest · Tier 2 — Notable_

![PrismML releases 5.9 GB Bonsai 2 model compressed from Qwen 27B — Primary](https://techcrunch.com/wp-content/uploads/2026/09/LLMs-on-smartphones.png?resize=1200,600)

PrismML on Thursday released Bonsai 2 27B, compressing Alibaba's open-source Qwen3.8 27B to 5.9 GB of memory, a 9x to 10x reduction the company says is small enough for a PC and possibly a high-end smartphone.

The Caltech-founded startup said the model matches 98% of Qwen's aggregate benchmark scores, up from 95% for the earlier Bonsai, which it said has been downloaded more than 11 million times. PrismML's method stores weights as +1, -1, or 0 instead of 16 bits. It has raised a $22.25 million seed round from Khosla Ventures, Cerberus Capital, and Caltech.

CEO Babak Hassibi said upcoming models will target several-hundred-billion parameters, where he expects compression to retain more intelligence. Advisor Ion Stoica said on-device models can run privately without cloud calls.

## Sources

- [Startups | TechCrunch](https://techcrunch.com/2026/09/17/prismml-hopes-its-tiny-llm-could-change-how-we-all-use-ai/)

---
Canonical: https://techandbusiness.org/newswire/SBcJrZp7i3kJjTRAIe0ELZ
Published: 2026-09-18T00:10:37.138Z
Story chronology: 2026-09-17T22:34:09.000Z
Retrieved: 2026-09-18T09:40:44.049Z
Publisher: Tech & Business (techandbusiness.org)