跳到正文
RadarAI· X:面壁智能 OpenBMB (@OpenBMB)·· 20 小时前AI 评分62

实测演示 MiniCPM5-2B 在本地完成多步智能体调查任务

AI 导读

@DailyDoseOfDS_ 实测让 OpenBMB 的 2B 稠密模型 MiniCPM5-2B 在 100% 本地环境驱动调查智能体,排查收入下滑原因。模型自行检查订单、流量、支付、退款和部署日志等文件,编写查询、分析中间结果并决定下一步,最终产出量化诊断和事件报告,数据、工具执行与模型推理均留在本地。

正文 · 原文

OpenBMB· @OpenBMB · X·2026-11-17 09:43· 1 小时前AI 评分64

AI 导读

@DailyDoseOfDS_ 分享了 100% 本地运行 OpenBMB 的 MiniCPM5-2B(2B 参数稠密模型)的实测,让它驱动一个调查智能体排查收入下滑原因。模型自行检查文件、编写查询、分析中间结果并决定下一步,最终产出量化的诊断和事件报告,数据、工具执行和模型推理全部留在本地。作者称其编码与工具调用能力来自 Agentic Pre-training、SFT 和大规模 RL,模型支持 SGLang、vLLM、llama.cpp、Ollama 及 iOS、Android、HarmonyOS,OpenBMB 已开源权重及部分训练配方和数据资源。OpenBMB 转发并感谢了这次演示。

正文 · 原文

Amazing demo! 🚀 Love seeing MiniCPM5-2B handle real-world, multi-step agent tasks entirely locally. Thanks for showcasing what small models can achieve! 💙

引用Daily Dose of Data Science@DailyDoseOfDS_

Insane progress for small language models!

MiniCPM5-2B is a dense 2B-parameter model by OpenBMB from China that is built for reasoning, coding, and tool use on resource-constrained hardware.

The model specifically excels at coding and tool calling, two capabilities central to the shift from on-device LLMs to on-device agents.

Instead of only answering prompts, it can use tools, generate code, carry information between steps, and complete multi-step tasks.

We ran it 100% locally and connected it to a small investigation agent with one request:

Revenue dropped last week. Investigate what happened, quantify the impact, identify the likely cause, and produce an incident report with supporting evidence.

The evidence was spread across orders, traffic, payments, refunds, and deployment logs.

The model inspected the files, wrote its own queries, analyzed the intermediate results, and decided what to investigate next.

Each tool result informed the next action, so the final report depended on the model maintaining a coherent investigation across the complete trajectory.

The recording shows the actual task from beginning to end. It starts with the revenue question, follows the tool calls and supporting evidence, and ends with a quantified diagnosis and incident report.

The data, tool execution, and model inference all remained on our machine.

These capabilities were optimized through Agentic Pre-training, SFT, and large-scale RL. They do not come entirely from an application-level agent framework.

MiniCPM5-2B supports SGLang, vLLM, llama.cpp, Ollama, iOS, Android, and HarmonyOS.

OpenBMB has also released the model weights and parts of the training recipes and data resources behind it.

A 2B model can now maintain enough state to coordinate tools and complete a useful investigation on local hardware.

在 X 查看被引用的帖子

来源:OpenBMB · x.com

来源:RadarAI · aihot.news