jevIMAGE
Jev 在 CUASpeedrun 电脑操作基准上被实测翻车
有人把 Jev 放进 cuaspeedrun.com 跑 computer use：比 Opus-5.5-xHigh 慢 1.5 倍、准确率低 10 倍、步数多 5 倍，只到 GPT-4 水平。
Pranjal Aggarwal ✈️ COLM'26
@PranjalAggarw162026.10.03收录原推发布于 2026.10.03
Pranjal Aggarwal ✈️ COLM'26 @PranjalAggarw16原推发布于 2026.10.03
Thought Jev is fast? We evaluated Jev on https://cuaspeedrun.com for computer use agents and found: 1. 1.5x slower than Opus-5.5-xHigh 2. 10x lower accuracy 3. 5x more steps That puts it in the GPT-4 league of performance and Astra-xHigh on time What and Why?? 🤯 🧵
x.com/pranjalaggarw16/status/2106091065388261599采集于 2026.10.03 10:05