PrismML ships Bonsai 2 27B: 9x smaller ternary model keeps 98.2% of full-precision performance
thione · x · 2026-09-21
PrismML released Ternary Bonsai 2 27B, built on Qwen3.8 27B with {-1, 0, +1} ternary weights and FP16 group-wise scaling, at an effective 1.76 bits per weight and a 5.9GB footprint, under Apache 2.0.
Key specs:
- Over 9x smaller than its full-precision counterpart while retaining 98.2% of aggregate benchmark performance;
- 262K-token context window with multimodal text-and-image input;
- Designed for local deployment: small memory footprint, high local throughput, better energy efficiency;
- Improved reasoning, coding, vision, and long-horizon agentic performance over the first Bonsai 27B.
Related event: PrismML's Ternary Bonsai 2 27B Shrinks Size 9x, Keeps 98.2% Performance(2 posts)→
More from Infra
- Meta partners with Arm on Arm AGI CPU, its first AI-era data center CPU — bookwormengr · 2026-09-21
- Starlink is becoming core infrastructure for rural education across Latin America — XFreeze · 2026-09-21
- HF speech-to-speech merges NVIDIA Parakeet Unified STT backend — andimarafioti · 2026-09-21
- Apple wins the AI infra lottery: M5 Ultra packs 512GB unified memory for local AI — Hesamation · 2026-09-21
- Z Image Turbo runs locally on a Snapdragon 8 Elite phone at 40s/image — Fine_Philosopher_882 · 2026-09-21
- Linux on 8GB RAM: you can run LLMs and do real development — gnukeith · 2026-09-21