Step 3.7 Flash 登顶 Claw-Eval General 第二
StepFun 的 Step 3.7 Flash 在智能体基准 Claw-Eval General 排第二,仅次于 Claude Opus 4.6,多步执行和长程任务都强,感兴趣可以看看。
StepFun 的 Step 3.7 Flash 模型在 Claw-Eval General 基准测试中取得第二名的成绩,该基准用于评估自主智能体。
模型在多步执行和长程任务鲁棒性上表现强劲,排名仅次于 Claude Opus 4.6。
这一结果显示其在真实世界智能体工作负载中的潜力。
Step 3.7 Flash hits #2 on Claw-Eval General for autonomous agents.
We’re seeing strong performance across multi-step execution and robustness in long-horizon tasks, ranking just behind Claude Opus 4.6.
Promising signals for real-world agent workloads. https://t.co/TghBHrrOym
This report is categorized under AITOP's "Models" desk. Key entities and technical topics under continuous tracking include Step 3.7 Flash, Claw-Eval General, agents, reasoning models.
Source attribution: curated from 阶跃星辰 Stepfun (@Stepfun_AI).