Speechmatics - API for Voice AI Agents
Mini tools1 Bit Bonsai 4B
By PrismML
1-bit Bonsai 4B is PrismML’s middle-tier model in the Bonsai line, built for workloads that need more capability than tiny models while staying extremely lightweight. PrismML says it requires only 0.57GB of memory and delivers 132 tokens per second on an M4 Pro, emphasizing a mix of strong accuracy, high speed, and energy efficiency. In practice, it is positioned as a local or edge-friendly model for developers who want a compact model that remains responsive enough for real-time assistants, embedded agents, and other latency-sensitive applications.
New
Text
Gen 7
Overview
1-bit Bonsai 4B is PrismML’s mid-size ultra-efficient model designed for fast local inference. It uses just 0.57GB of memory and is positioned as a strong balance of speed, accuracy, and energy efficiency, with PrismML reporting throughput of 132 tokens per second on an M4 Pro for workloads that need both responsiveness and capable language performance.
About PrismML
Industry:
Artificial Intelligence
Company Size:
6
Location:
Pasadena, California, US
Website:
prismml.com
Tools using 1 Bit Bonsai 4B
No tools found for this model yet.
Last updated: April 1, 2026
