面壁智能回应 MiniCPM5 小模型实测对比
Love seeing these real-world comparisons! 💙 Thanks for including MiniCPM5 in your tests. Exciting to see how far small models have come! 🚀
GooGZ AI 在 RTX 3080 上以 PA Core 测试三款 2B 以下小模型，称 Spark X2.6 1.7B、LFM3 与 MiniCPM5 各有强项，其中 MiniCPM5 表现最均衡。面壁智能 OpenBMB 转发该测试并致谢，表示很高兴看到小模型进步之大。
查看原帖小模型眼下正经历一轮真正的快速增长。 在我的个人智能体测试中（PA Core，开启推理，RTX 3080），目前三款 2B 及以下的模型彼此分差在 21 分以内，并且与更早的 4B/9B 权重模型不相上下。 最新加入测试的是来自 @SparkLLM 的 Spark X2.6 1.7B - PA 总分：172.3/300 - 工具调用：每次运行均为 30/30 - 上下文召回：8K 时为 9.3，16K 时为 8.7，是三款中最好的！ - 日程与草稿（PA2）：两次运行分别为 98 和 100，一次 54 把均值拉低到 84 - 日常整理（PA1）：35.7，是薄弱环节 在 2B 以下，三款各有强项。@liquidai 的 LFM3 擅长规划，@SparkLLM 的 Spark 擅长读取，@OpenBMB 的 MiniCPM5 则最为均衡。 保持思考开启。若关闭思考，Spark 在这些任务上会弱得多。
查看原文
Small models are going through a real growth spurt right now. On my personal-agent tests (PA Core, reasoning on, RTX 3080), three models at 2B and under now land within 21 points of each other, and rubbing shoulders with older 4B/9B weights. Newest on the bench: Spark X2.6 1.7B from @SparkLLM - PA total: 172.3/300 - Tool calls: 30/30 in every run - Context recall: 9.3 at 8K, 8.7 at 16K, best of the three! - Schedule and draft (PA2): 98 and 100 in two runs, one 54 pulled the mean to 84 - Daily organization (PA1): 35.7, the weak spot Under 2B, three different strengths. @liquidai LFM3 plans, @SparkLLM Spark reads, @OpenBMB MiniCPM5 is the balanced one. Leave thinking on. With it off, Spark is a much weaker model on these tasks.
正文
Love seeing these real-world comparisons! 💙 Thanks for including MiniCPM5 in your tests. Exciting to see how far small models have come! 🚀