RadarAI· X：面壁智能 OpenBMB (@OpenBMB)·· 5 天前AI 评分34
MiniCPM-V-4.7-35B-A3B 实测智能体工作流
AI 导读
OpenBMB 的 MiniCPM-V-4.7-35B-A3B 在真实智能体工作流测试中表现突出，在个人智能体基准中超越同级 35B MoE 模型，Lite Agent 测试（A1）取得完美结果，PA4 复杂智能体循环工作流（含大量上下文与工具调用）也表现强劲。该测试尚未计入其新增视觉等能力。
正文 · 原文
OpenBMB· @OpenBMB · X·2026-10-13 11:36· 1 小时前AI 评分26
AI 导读
很高兴看到 MiniCPM-V-4.7-35B-A3B 在真实智能体工作流中接受测试。👀 强大的工具调用、长上下文处理以及智能体循环，正是高效模型需要证明自己的地方。
正文
Great to see MiniCPM-V-4.7-35B-A3B being tested in real agent workflows. 👀
Strong tool use, long context handling, and agent loops are exactly where efficient models need to prove themselves.
I appreciated @OpenBMB giving a few of us an early look at these new models. They are the real deal. When I put MiniCPM-V 4.7 35B A3B through my own personal agent benchmarking (see comments for more info on what I test and why), it outperforms 35B MoE incumbents that one would normally reach for. Lite Agent testing (A1) was a perfect result, and PA4 (bringing a complex agent loop workflow together with lots of context and tool calls) was surprisingly strong. This is even before looking at the new vision and other capabilities, but as a life admin 'brain' - it's quite stellar.在 X 查看被引用的帖子
来源：OpenBMB · x.com
来源：RadarAI · aihot.news