返回探索

Jev与Gemini对比测试:300个客户请求

作者在300个客户请求上测试Jev和Gemini,发现提出更多问题可将模型调用从约7次减少到3次,并提升两者准确性。Jev在过滤错误答案方面表现出近4倍的更好权衡。

1/7 I tested Jev and Gemini on 300 customer requests. Asking MORE questions cut the agent from ~7 model calls to 3, and made BOTH models more accurate. But Jev also had a nearly 4x better trade-off when filtering wrong answers. Then Gemini hit a limit that had nothing to do

Jev与Gemini对比测试:300个客户请求 1
· 1 次赞在 X 打开