StepFun发布Step 5 Preview模型，编程与长上下文能力获验证
StepFun’s new Step 5 Preview model is impressive!
StepFun’s new Step 5 Preview model is impressive!
StepFun 的新 Step 5 Preview 模型令人印象深刻！
Had a chance to test it early.
我有机会提前测试了它。
I've been testing it as a coding agent.
我一直将其作为编码代理进行测试。
It’s on the Pareto frontier for cost vs. capability.
它在成本与能力的权衡上处于帕累托前沿。
It is a very capable model, comparable to GLM 5.3, Kimi K3, and others, and it comes at a very competitive price.
它是一个能力非常强的模型，可与 GLM 5.3、Kimi K3 等相媲美，且价格极具竞争力。
It’s worth trying in your favorite coding agent.
值得在你常用的编码代理中尝试一下。
I tested it in a minimal harness to see how it compares with GLM 5.3.
我在一个最小化测试框架中测试了它，以观察它与 GLM 5.3 的对比情况。
Here are my results.
以下是我的测试结果。
Overall, it is very effective and feels like a model I could use for a whole range of coding tasks. One behavior that stood out is that it knows when to stop, which makes it very effective at long-horizon tasks compared with other models in this class.
总体而言，它非常高效，感觉像是一个可以用于各种编码任务的模型。其中一个突出的行为是它知道何时停止，这使得它在长周期任务中比同类其他模型更加有效。
I gave it and GLM 5.3 two real tasks in the same repo, at the same commit. First, a bug where numeric filters silently returned zero rows for decimals and negatives. Then a feature that needed new routes, permission gating, and a refactor of the background task supervisor.
我让它在同一个仓库、同一个提交版本下处理了两个真实任务。首先是一个 bug：数值过滤器在遇到小数和负数时会静默返回零行。其次是一个需要新增路由、权限门控以及重构后台任务主管的功能。
Both models got both tasks right. Every held-out test passed with no regressions, and neither one weakened an existing test to get there.
两个模型都正确完成了这两项任务。所有预留测试均通过，无回归问题，且两者都没有通过削弱现有测试来达成目标。
Step 5 Preview finished, checked its work, and declared itself done. Both times.
Step 5 Preview 完成了任务，检查了自己的工作，并宣布自己已完成。两次都是如此。
GLM 5.3 wrote correct code both times and then kept going until the step limit ended the run.
GLM 5.3 两次都写出了正确的代码，然后继续运行直到达到步骤限制而终止。
Neither model got a follow-up prompt or a retry. Both had zero fix rounds, so the first delivery was the final delivery on both tasks.
两个模型都没有收到后续提示或重试指令。两者的修复轮次均为零，因此首次交付即为最终交付。
On the bug, Step 5 Preview wrote the shorter patch, the same approach the Datasette maintainer used in the real commit. It also added its own tests without being asked.
在处理 bug 时，Step 5 Preview 写出了更短的补丁，采用了与 Datasette 维护者在实际提交中相同的方案。它还主动添加了测试，尽管并未被要求这样做。
Based on this, I would reach for it on unattended agent runs where a clear completion signal matters more than speed, on bug fixes in unfamiliar codebases, and on long-context work.
基于此，我认为在无人工干预的代理运行中，当明确的完成信号比速度更重要时，或者在处理不熟悉代码库中的 bug 修复以及长上下文工作时，我会优先选择使用它。
On long context, I gave it a separate test earlier in the week. I generated about 368K tokens of fake incident tickets and hid five clues inside them that together explain an outage. When I asked for the root cause, it found all five clues and connected them correctly in about 90 seconds.
关于长上下文处理，我在本周早些时候给它安排了单独的测试。我生成了约 368K token 的虚假故障工单，并在其中隐藏了五个线索，这些线索共同解释了一次停机事故。当我询问根本原因时，它在大约 90 秒内找到了所有五个线索并正确地将它们关联起来。
Thanks to the StepFun team for partnering on this post.
感谢 StepFun 团队对本帖的合作支持。
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力