New todaydeepseek-v4.1POST

DeepSeek V4.1 Flash catches 48 of 60 planted bugs, beating Sonnet 5.5

With one bug planted in each of 60 real open-source diffs, Opus 5.5 and Gemini 4 Argon scored 54, GPT-6.1 Sol 49, DeepSeek V4.1 Flash 48 and Sonnet 5.5 42.

Alex Isaev

@alexisaevArchived 11 hours agoOriginally posted 2026.10.28

Alex Isaev @alexisaevOriginally posted 2026.10.28

Open original on X

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/2115269984553467968Archived 2026.10.28 12:01