Moonshot AI 发布 Kimi K3.1 Pro：Terminal-Bench 得 65.7%，11 月 4 日开放全量权重
月之暗面更新了 Kimi K3.1 Pro，跑分紧咬 Claude Opus 5.5，成本还更低，11 月 4 日放出全量权重可以自己复测。
Moonshot AI 发布 Kimi K3.1 Pro，在 Terminal-Bench 4.0 上拿到 65.7%，比 Claude Opus 5.5 低 0.7 分，比 GPT-6 Astra 高 7.5 分，单任务平均成本低于两者。为抑制漏洞利用能力，Moonshot 在中期训练和 RL 阶段移除了网络安全数据，内部 xdev-bench-hard 的 pass@3 从 90.5% 降到 81.0%。工具调用幻觉测试中，虚假陈述比例从 K3 的 23.4% 降到 8.3%，但 Terminal-Bench Science 仅得 38.1，落后 GPT-6 Astra 的 68.1。K3.1 Pro 和 2-3 倍速的 Highspeed 版本已上线 Kimi 应用与 API，全量权重将在 11 月 4 日放出。
Moonshot AI’s first update, the new Kimi K3.1 Pro
> Kimi K3.1 Pro scores 65.7% on Terminal-Bench 4.0, 0.7 points behind Claude Opus 5.5 and 7.5 ahead of GPT-6 Astra, at a lower average cost per task than either rival’s top runs.
> Moonshot saw exploit skill climb as post-training scaled, so it removed cyber-related data from mid-training and RL; on its internal xdev-bench-hard, pass@3 fell from 90.5% to 81.0%.
> The filtered model still solves 81.0% of those exploitation tasks against K3’s 33.3%, which Moonshot reads as general coding gains carrying into cyber work, while a third-party run of GPT-5.6 Sol solved 100%.
> On Moonshot’s internal hallucination stress test with tools, false claims fell from 23.4% for K3 to 8.3%, still above GPT-6 Astra’s 5.5% but well below Claude Opus 5.5’s 15.0%.
> The biggest jump is on Moonshot’s simulated business-operation test, where scores rose from 8.9 to 37, and K3.1 Pro also tops OfficeQA Pro v2 at 64.4 against Opus 5.5’s 62.2.
> Gains are uneven: K3.1 Pro scores 38.1 on Terminal-Bench Science against GPT-6 Astra’s 68.1 and Opus 5.5’s 63.3, while leading Agents’ Last Exam at 43.2 and Humanity’s Last Exam with tools at 70.0.
> K3.1 Pro and a Highspeed variant running 2-3x faster are live in Kimi’s apps and API, with full weights due by November 4 for independent reruns.
