OpenBMB 发布开源多模态模型 MiniCPM-V 4.7 系列
OpenBMB 的 MiniCPM-V 4.7 能在 iPhone 上跑 35B 多模态模型，3GB 内存 13 tokens/s，OCR 还压过 Qwen3.6，端侧玩家可以试试。
OpenBMB 推出开源多模态模型 MiniCPM-V 4.7，包含 35B-A3B 和 1B 两个版本，可在 iOS 和 Android 设备本地运行。35B-A3B 版本采用 LES（Lookahead Expert Scheduling）技术，在 iPad Pro M4（16GB）上实现 17+ tokens/s、iPhone 17 Pro Max 上 13+ tokens/s 的推理速度，内存占用仅 3GB。模型在 OCR、STEM 推理、长视频理解等 34 项基准上取得有竞争力的成绩，16 倍视觉 token 压缩下 OCRBench 得分 90.4，比 Qwen3.6 35B-A3B 高 5.2 分。新版本还提供 Self-Replay Fine-Tuning 微调方法，并首发适配 Apple、Intel、AMD、Arm 和 Rockchip 硬件。
🚀 Introducing MiniCPM-V 4.7, our new open-source MLLM series! Advancing the Performance–Efficiency Pareto Frontier. It brings strong multimodal capabilities to iOS and Android devices with extreme efficiency. Two new models: 🔹 MiniCPM-V 4.7 35B-A3B — Bring SOTA performance to local devices with newly introduced LES (Lookahead Expert Scheduling) technique, a new milestone of on-device multimodal AI. 🔹 MiniCPM-V 4.7 1B — Lightweight, efficient multimodal AI. Easier to deploy.
⚡ What's new? 📊 SOTA performance: Competitive results across 34 benchmarks spanning multimodal understanding, STEM reasoning, OCR and long-video comprehension. ⚡Fast on-device inference: LES predicts and preloads expert weights to reduce inference stalls, boosting 5× speedup on iPad Pro M4 (16GB) with byte-for-byte identical outputs. It enables MiniCPM-V 4.7 35B-A3B to achieve 17+ tokens/s on iPad Pro M4 (16GB) and 13+ tokens/s on iPhone 17 Pro Max, while costing only 3GB memory 🤯 🧩 Efficient visual processing: 4× and 16× hybrid visual token compression. Even at 16× compression, 35B-A3B scores 90.4 on OCRBench, 5.2 points ahead of Qwen3.6 35B-A3B. 🛠️ Easier customization: We introduce Self-Replay Fine-Tuning, a simple way to customize new target capabilities at low cost while maintaining general performance. 💻 Broad hardware support: Day-0 adaptations for Apple, Intel, AMD, Arm and Rockchip, with many more platforms supported through FlagOS!
Download, deploy, and build something we haven't imagined yet. 👀👇 🔗 GitHub: https://t.co/msjm2QlrvX 🤗 Hugging Face: https://t.co/W95QKGhymw
- Gorden Sun10:24原文
