Description

Bonsai 2 27B uses 'ternary weights' to compress the Qwen3.8-27B model down to a 5.9 GB footprint, making it 9 times smaller while maintaining nearly the same benchmark performance. By limiting its weight values to only -1, 0, or +1, it significantly improves memory efficiency and allows high-capability models to run on standard consumer GPUs.

Descriptions, tags, and model credits may be AI-generated or inferred from public sources and can be incomplete or wrong. Learn more.