今天新增deepseek-v4.1POST
60 个真实开源 diff 植入 bug:DeepSeek V4.1 Flash 修出 48 个,超 Sonnet 5.
在 flask、zod、svelte 等 60 个真实 diff 各植入一个 bug,Opus 5.5 与 Gemini 4 Argon 各 54、GPT-6.1 Sol 49、DeepSeek V4.1 Flash 48、Sonnet 5.5 42。
Alex Isaev
@alexisaev7 小时前收录原推发布于 2026.10.28
Alex Isaev @alexisaev原推发布于 2026.10.28
deepseek's flash model caught 6 more planted bugs than sonnet 5.5 we planted one bug in each of 60 real open-source diffs (flask, zod, svelte...) opus 5.5: 54 gemini 4 argon: 54 gpt-6.1 sol: 49 deepseek v4.1 flash: 48 sonnet 5.5: 42 private tests are where leaderboards get weird
x.com/alexisaev/status/2115269984553467968采集于 2026.10.28 12:01