Bonsai-27B-mlx-1bit is an aggressively quantized 27B parameter multimodal model designed to run efficiently on Apple Silicon using the MLX framework. It supports vision and text inputs while using extremely low bit precision. The model is distributed via Hugging Face for local deployment and experimentation.
In the Foundation models & chat space, Bonsai 27B Mlx 1bit takes a focused approach. It focuses on running very large models with minimal memory on local devices. It is built as an open-source project for developers. Bonsai 27B Mlx 1bit is open source under the Apache-2.0 license. Bonsai 27B Mlx 1bit is available on the web, the command line, and API.
Behind Bonsai 27B Mlx 1bit is prism-ml, and it first shipped in 2026. Development happens publicly on GitHub with 2.1k stars and 35 commits in the last 90 days. It exposes integrations via a public API.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do