Event ·Multi-source

Kimi-K3.1-Pro 发布：3T 参数，SWE 系列测试对标 OpenAI 与 Anthropic 旗舰

月之暗面发布 Kimi-K3.1-Pro，参数规模达 3T。官方跑分显示其代码能力介于 Claude Opus-5.5 与 Fable-5.1 之间，SWE 系列 Agentic Coding 测试成绩与 OpenAI、Anthropic 旗舰模型相当，TerminalBench 成绩创下新高。该模型还引入了新基准 BencCAD 用于测试 CAD 建模能力，官方博客附带了多个 CAD 建模演示。定价信息尚未公布。

Current read

Moonshot 新发的 Kimi-K3.1-Pro 有 3T 参数，写代码跑分能跟 OpenAI、Anthropic 旗舰掰手腕，还带了个测 CAD 的新基准，演示挺炫。

11 sources79° AI heatLast updated 2026/10/23 17:33:25

Evidence

11 sources

Primary sourcekarminski-牙医 (AI工具)

2026/10/23 17:33:25View source

Related source 1Decoder

2026/10/23 19:00:07View source

Related source 2Artificial Analysis

2026/10/22 02:04:56View source

Related source 3TestingCatalog

2026/10/23 07:55:08View source

Related source 4kimmonismus

2026/10/23 17:18:42View source

Related source 5Bindu Reddy

2026/10/23 10:57:04View source

Related source 6ARC Prize

2026/10/21 19:30:47View source

Related source 7rohanpaul_ai

2026/10/21 19:48:33View source

Related source 8IT之家

2026/10/21 22:44:18View source

Related source 9Simon Willison’s Weblog

2026/10/22 18:09:56View source

Related source 10Blognone

2026/10/23 00:31:01View 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