New models combine reasoning, vision, and tool calling to support AI agents on everyday devices
PASADENA, Calif., Oct. 15, 2026 /PRNewswire/ — PrismML today expanded its Bonsai 2 family with new 2B, 4B, and 9B ternary models based on Qwen3.5. The releases bring stronger capabilities to smaller deployment footprints designed for the resource limits of personal computers, phones, smartglasses, and other wearables. The Ternary Bonsai 2 family now includes four sizes spanning 2B to 27B parameters.
Building on the first-generation Bonsai 1.7B, 4B and 8B models, the new releases combine improved reasoning, visual understanding, and support for tool-driven workflows. The models use PrismML’s ternary weight representation, which encodes model weights using three values: {−1, 0, +1}. The 2B, 4B and 9B models have footprints of 0.43GB, 0.95GB and 2.00GB, respectively, more than 9× smaller than their full-precision counterparts.
“On-device agents have to be smart enough to make the right decisions while working within limited memory and power budgets,” said Babak Hassibi, Founder and CEO of PrismML. “With these Ternary Bonsai 2 releases, we’re giving developers more capability within those constraints and more flexibility in what they can build.”
Stronger performance for local AI
Across a suite of 19 benchmarks, the new Ternary Bonsai 2 models match or exceed the performance of their full precision counterparts on coding, math, and instruction following tasks. On tasks related to vision, agentic work and tool-calling, and knowledge and reasoning, the models retain well over 90% of the benchmark scores of the full precision Qwen3.5 models. In all cases, the Ternary Bonsai 2 models show considerable performance gains over their first-generation Ternary Bonsai counterparts.
For agents that alternate between interpreting information, calling tools and generating responses, faster inference can reduce the time spent preparing each action or response.
On hardware such as RTX 4090 and Macbook M5 Pro, Bonsai 2 models deliver up to 3x the energy efficiency and 5x the throughput of their corresponding full-precision models. Bonsai 2 9B reaches up to 124 toks/s, while Bonsai 2 2B reaches up to 276 toks/s.
Their compact footprints are also ideal for mobile deployments. On an iPhone 18 Pro Max, Bonsai 2 9B delivers up to 34 toks/s tokens per second, while Bonsai 2 2B reaches up to 120 toks/s.
Building applications with Bonsai 2
At the Snapdragon Summit in September, PrismML demonstrated a 1-bit Bonsai vision-language model running locally on AI smart glasses powered by the Snapdragon AR1 Gen 1 Platform. Built on first-generation Bonsai, the model showed how visual understanding and assistance could operate within the memory and power constraints of wearable hardware.
The new Bonsai 2 models build on that progress with stronger reasoning, visual understanding, and tool-calling capabilities. These capabilities provide a foundation for bringing basic agentic functionality to future applications on phones, smart glasses, and other wearables. Developers can build assistants that interpret a user’s request, reason over text or visual context, and use application tools to carry out short sequences of actions.
Technical Details:
The Ternary Bonsai 2 2B, 4B and 9B models have been trained using Google v5 TPUs. They have been designed for seamless integration with existing AI workflows and are optimized for low-latency inference on consumer-grade CPUs, and edge GPUs.
Pricing and Availability:
Developers, researchers, and other users can download the new Bonsai 2 models under the Apache 2.0 license for free starting October 15, 2026.
Download Ternary Bonsai 2 9B, 4B, 2B model here
About PrismML:
PrismML is a U.S.-based artificial intelligence company focused on making AI more efficient and accessible. PrismML emerged from a team of Caltech researchers and was founded with support from Khosla Ventures, Cerberus, and Google, with continuing support from Samsung. For more information, visit the Website, LinkedIn, or X.
All registered trademarks and product identifiers belong to their respective corporate entities. Any other trademarks or product names referenced here are also owned exclusively by their relevant companies.
Media Contact:
PrismML
Gary Bird
gary@prismml.com
831.888.9011
View original content:https://www.prnewswire.com/news-releases/prismml-expands-bonsai-2-family-with-2b-4b-and-9b-models-for-on-device-ai-302910593.html
SOURCE PrismML