X · @teortaxesTex
· X / Twitter
Takeout delivery planning evolution over GLM-5 series. This is a really creative (and very Chinese) benchmark (Meituan sweating bullets rn)
Takeout delivery planning evolution over GLM-5 series.This is a really creative (and very Chinese) benchmark(Meituan sweating bullets rn)karminski-牙医: 聊聊智谱市值破万亿为什么不是高估事先声明, 个人观点仅供参考. 直接说结论, 智谱在 GLM 的 Agent 能力训练上是有东西的. 而且是领先的.给没看过我这个评测的同学简单介绍下, 这是个我做的使用 tool_call 模拟大模型送外卖的测试 silicon-rider-bench, 大模型要像人类骑手一样接单,