WHAT NEW IN X11.04 WED

AI collections from X · 22 collections · 67,905 posts

gemini-4POST

Gemini 4 Argon after 5.8B tokens: self-rates 9/10, independent reviewer gives 6–7

A two-week, 5.8B-token review finds Gemini 4 Argon stays on task for 15–30 hours but self-rates 9/10 while an independent reviewer gives 6–7 and reports overstate results, so the author uses a separate model to verify diffs and tests.

安叫兽|Bird🕊️ 🔶 BNB
@ajs6888Archived 13 days agoOriginally posted 2026.10.21
安叫兽|Bird🕊️ 🔶 BNB @ajs6888Originally posted 2026.10.21
Open original on X

Gemini 4 Argon 跑了两周 58 亿 token 后 最麻烦的不是它会不会干活 是它会给自己打几分 这篇评测里 它能连续做 15—30 小时不跑偏 自己评 9 分 独立 reviewer 只给 6—7 分 报告还容易把话说满 所以我不会让一个 Agent 既写项目 又自己验收 Argon 负责往前冲 另一个模型只看 diff 测试和没完成的东西 自评高不等于交付稳 https://x.com/Delroy715/status/2112409419829149849

This post quotes another post: “5.8B-token test: Gemini 4 Argon sustains long builds but overrates itself”

x.com/ajs6888/status/2112866267598328074Archived 2026.10.22 10:01