手机能跑 27B 模型了？PrismML 的 Bonsai 27B 实际做到了——体积最小 3.9GB，速度还不慢，开源免费。想本地用大模型又嫌大的可以看。
PrismML 今日发布 Bonsai 27B，这是首个可在手机上本地运行的 27B 参数多模态模型。基于 Qwen3.6 27B，Bonsai 27B 支持多步推理、结构化工具使用、长上下文和智能体循环。Ternary 变体仅 5.9 GB（1.71 有效比特/权重），1-bit 变体仅 3.9 GB（1.125 有效比特/权重），大幅降低本地部署门槛。在 RTX 5090 上，1-bit 版本达到 163 tok/s，Ternary 版本达到 134 tok/s；在 M5 Max 上分别为 87 tok/s 和 58 tok/s。模型以 Apache 2.0 许可证开源。
Huge if true! We are talking about a 27B multimodal model that runs locally on a phone. That's wild...
Huge if true! We are talking about a 27B multimodal model that runs locally on a phone. That's wild! Bonsai 27B reaches up to 163 tok/s in 1-bit and 134 tok/s in Ternary on an NVIDIA GeForce RTX 5090. On an M5 Max, it reaches up to 87 tok/s in 1-bit and 58 tok/s in Ternary. PrismML @PrismML Today, we’re announcing Bonsai 27B: the first 27B-class model to run on a phone. Bonsai 27B is the new multimodal flagship of the Bonsai family. Based on Qwen3.6 27B, it brings a new capability tier to local AI: multi-step reasoning, structured tool use, long-context workflows, and coherent agentic loops. Until now, models in this class have been impractical to deploy locally. A 27B model occupies roughly 54 GB in 16-bit precision, and even a strong 4-bit build is around 18GB - too large for a phone and for most laptops. Bonsai 27B changes that. It comes in two variants: • Ternary Bonsai 27B: 5.9 GB, 1.71 effective bits per weight, optimized for laptop-class quality. • 1-bit Bonsai 27B: 3.9 GB, 1.125 effective bits per weight, optimized for phone-class footprint. Everything is open-sourced today under the Apache 2.0 license. 🔗 View Quoted Tweet 💬 7 🔄 4 ❤️ 34 👀 8360 📊 11 ⚡