Benchmarking Opus 5 on SlopCodeBench

在 SlopCodeBench 上基准测试 Opus 5

HN 188 分 · 46 条评论 · 作者 dhorthy · 来源 github.com · HN 讨论

【摘要】

The article analyzes SlopCodeBench, a long-horizon coding benchmark that reveals requirements gradually through multiple checkpoints rather than upfront. This design tests a model's ability to evolve a codebase over time as new constraints are introduced.

⋯ 继续阅读请登录会员 ⋯