Ternary Bonsai-2 27B reads medication lists locally on Mac in 5.7 seconds
MaziyarPanahi · x · 2026-09-18
Prism ML released the Bonsai-2 collection on Hugging Face, centered on a 27B ternary model available in GGUF, MLX 2-bit, and ONNX formats, with a WebGPU kernel that runs the model directly in the browser.
The author demonstrated Bonsai 2 running locally on a Mac Studio: from just two photos of a medication list, it spotted a dose change from 5mg to 10mg and a missing drug in the latest list, completing the run in about 5.7 seconds — a showcase of practical, low-cost local AI.
Related event: PrismML's Ternary Bonsai 2 27B Shrinks to 5.9GB, Retains 98.2% Performance(20 posts)→
More from Infra
- Training-free page-sparse decode in llama.cpp: 1.72× faster 128k decoding on an RTX 5090 — topppggg · 2026-11-16
- AMD MI455 beats Nvidia Rubin on raw specs, claims viral chip debate — zephyr_z9 · 2026-11-16
- llama.cpp offloads CI caches and runners to Hugging Face to survive 35 PRs/day — ggerganov · 2026-11-16
- India's Sarvam launches Token Factory, hosting top open models on local infra — HarveenChadha · 2026-11-16
- ZGX Fury stress-tested: 351 tok/s on Qwen3.8-Flash-Next, offline agents and local RAG demos — Sam Witteveen · 2026-11-16
- Gemini 3.5 Flash now runs on air-gapped Google Distributed Cloud, offline up to 12 months — shashib · 2026-11-16