{"slug":"ternary-bonsai-2-27b","name":"Ternary Bonsai 2 27B","nameZh":"Bonsai 2 27B（三值量化，约 6GB）","category":"models","categoryName":"Open-weight models","audience":["ai-era"],"primary":"ai-era","openClosed":"open","pricing":"open-source","license":"Apache-2.0","maintained":"yes","officialUrl":"https://huggingface.co/prism-ml/Ternary-Bonsai-2-27B-gguf","repo":null,"latestVersion":null,"latestRelease":null,"stars":null,"starsAsOf":null,"description":"Prism ML’s ternary (−1/0/+1) build of Qwen3.8-27B: ~1.72 bits per language weight, GGUF ~5.95–7.21GB, the card claims 98.2% of FP16 quality and 262K context; runs on its llama.cpp fork (CUDA / Metal / CPU), with an MLX build for Apple silicon. Apache-2.0","descriptionZh":"Prism ML 基于 Qwen3.8-27B 的三值（−1/0/+1）量化版：语言权重约 1.72 bit/权重，GGUF 约 5.95–7.21GB，模型卡称保留 FP16 98.2% 的能力、262K 上下文；需用其 llama.cpp 分支（CUDA / Metal / CPU），另有 MLX 版供 Apple 芯片。Apache-2.0","callMethods":[],"platforms":[],"addedAt":"2026-10-20","updatedAt":null,"topics":[],"language":null,"lastPush":null,"formerly":null,"packages":[],"recentVersions":[],"alternatives":["deepseek-v3","qwen3","gpt-oss","qwen3-coder","kimi-k2","llama-models","ultralytics","gemma"],"stacks":[],"scenarios":[],"problems":[],"reasonZh":null,"sources":["editorial"],"record":"https://aimap.page/api/v1/software/ternary-bonsai-2-27b.json","pages":{"en":"https://aimap.page/en/software/ternary-bonsai-2-27b/","zh-hans":"https://aimap.page/zh-hans/software/ternary-bonsai-2-27b/"}}