面壁智能 MiniCPM-V 4.7 上线 ModelScope 开源
Great to see MiniCPM-V 4.7 on ModelScope! 🚀
面壁智能宣布 MiniCPM-V 4.7 已在 ModelScope 开源，这是一个 35B-A3B MoE 多模态模型，采用 Apache 2.0 许可，主打端侧运行。据 ModelScope 介绍，该模型在 iPhone 17 Pro Max 上以 13+ tokens/s 运行，是首个支持流畅无损端侧推理的开源 MoE，并提供 35B-A3B 与 1B 两种尺寸及 FP8、GPTQ、AWQ、BNB、GGUF 等变体，兼容 SGLang、vLLM、llama.cpp、Ollama 等工具，支持 iOS、Android 和 HarmonyOS。
推荐理由
AI 评分 69
面壁智能开源 MiniCPM-V 4.7，给出端侧多模态模型在手机上的推理速度与压缩方案，可据此判断端侧部署的可行性。
主题
查看原帖MiniCPM-V 4.7 is now open source on ModelScope: a 35B-A3B MoE multimodal model built to run on-device without compromising performance. Apache 2.0. 🤖 https://t.co/NwwbEXRXGg 📱 First open-source MoE with smooth, lossless on-device inference, running at 13+ tokens/s on iPhone 17 Pro Max. 🔥 New Pareto frontier: in Instruct mode, it reaches a higher overall…
展开完整原文
MiniCPM-V 4.7 is now open source on ModelScope: a 35B-A3B MoE multimodal model built to run on-device without compromising performance. Apache 2.0. 🤖 https://t.co/NwwbEXRXGg 📱 First open-source MoE with smooth, lossless on-device inference, running at 13+ tokens/s on iPhone 17 Pro Max. 🔥 New Pareto frontier: in Instruct mode, it reaches a higher overall score in less inference time than Qwen3.6 35B-A3B. ⚡ Mixed 4x/16x visual token compression lets developers balance visual detail and inference cost per task. 🛠️ Two sizes (35B-A3B and 1B) with FP8, GPTQ, AWQ, BNB, and GGUF variants. Works with SGLang, vLLM, llama.cpp, Ollama, SWIFT, and LLaMA-Factory. 💻 Runs on iOS, Android, and HarmonyOS, with day-0 support for Rockchip, Houmo, Intel, AMD, and Arm.
正文
当前显示原文，中文全文译文尚未就绪。
Great to see MiniCPM-V 4.7 on ModelScope! 🚀
Thanks for the support 💙 Excited to make powerful on-device multimodal AI accessible to even more developers!