Event ·Official first-hand

Kimi K3投资研究能力获基准测试，击败GPT-5.5和OPUS-4.8

Moonshot AI旗下Kimi K3模型在AlphaEngine的专有IRB投资研究基准测试中，以更高得分击败GPT-5.5和OPUS-4.8。测试涵盖时间判断、证据边界控制和前提修正三个维度。Kimi K3已部署在AlphaEngine平台上。

Current read

Kimi K3在投资研究上直接干翻了GPT-5.5和OPUS-4.8，时间判断和证据边界控制尤其强，已经在AlphaEngine上用了，搞金融研究的可以试试。

11 sources47° AI heatLast updated 2026/7/20 07:57:42

Evidence

11 sources

Primary sourcePandaily

2026/7/20 07:57:42View source

Related source 1AI Will

2026/7/21 08:21:38View source

Related source 2Epoch AI

2026/7/21 17:41:46View source

Related source 3vLLM

2026/7/18 12:25:38View source

Related source 4Aravind Srinivas

2026/7/18 18:44:40View source

Related source 5techcrunch

2026/7/18 18:51:07View source

Related source 6Ethan Mollick

2026/7/18 23:58:03View source

Related source 7Guillermo Rauch

2026/7/19 01:06:33View source

Related source 8Decoder

2026/7/19 09:32:19View source

Related source 9@koltregaskes

2026/7/19 12:06:44View source

Related source 10shao__meng

2026/7/19 12:18:33View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.

Open the story