Gemini 4 Argon 在 105 个真实 bug 修复测试中排第四

Paweł Huryn @PawelHuryn原推发布于 2026.10.14

在 X 上打开原推

I tested Gemini 4 Argon on real work. 2 repos. 105 bugs frontier models missed in Q1 2026. Find and fix what you can. The results: GPT-6 Astra (max): 45 Fable 5.1 (max): 43 Opus 5.5 (max): 41.7 Gemini 4 Argon (high, its max): 33 Muse Spark 1.3 (max): 32.2 Grok 4.7 (xhigh, its max): 29 A way better model than Gemini 3.8 Flash, tested at its max. Based on the history, the uncertainty is 2-3 points. Looks like the Google is back.