本文为原文前 6,000 字符的节选翻译，完整内容请查看原文。

Kimi K3.1 Pro and K3.1 Pro Highspeed

Scroll to exploreIntroducing Kimi K3.1 ProWe’re introducing Kimi K3.1 Pro, the first model in our new K3.1 series and our most capable 3T-class open-weight model to date, with a 1-million-token context window. It delivers substantial improvements across coding, 3D creation, financial research, and other agentic capabilities. We are also releasing Kimi K3.1 Pro Highspeed, which delivers 2-3× the speed of K3.1 Pro for more responsive interactions. Kimi K3.1 Pro and Kimi K3.1 Pro Highspeed are available today on Kimi.ai, Kimi Work, Kimi Code, and the Kimi API. The full model weights will be released by Nov 4.

滚动探索 Kimi K3.1 Pro。我们推出了 Kimi K3.1 Pro，这是我们全新 K3.1 系列中的首款模型，也是迄今为止我们能力最强的 3T 级开源权重模型，拥有 100 万 token 的上下文窗口。它在编程、3D 创作、金融研究及其他智能体能力方面带来了显著提升。我们还发布了 Kimi K3.1 Pro Highspeed，其速度是 K3.1 Pro 的 2-3 倍，可实现更灵敏的交互。Kimi K3.1 Pro 和 Kimi K3.1 Pro Highspeed 即日起在 Kimi.ai、Kimi Work、Kimi Code 和 Kimi API 上线。完整模型权重将于 11 月 4 日发布。

Advancing the Frontier of Open Intelligence Kimi K3.1 Pro is a post-trained version of Kimi K3, with substantially larger post-training compute. To enable large-scale joint RL across hundreds of downstream domains, we optimized every layer of the stack — the RL framework, long-context inference, training efficiency, sandbox infrastructure, rollout scheduling, and a shared platform for environment quality checks — so that each increment of RL compute delivers stable, predictable returns. As a result, K3.1 Pro achieves consistent gains in every domain.

推进开放智能的前沿。Kimi K3.1 Pro 是 Kimi K3 的后训练版本，拥有大幅增加的后训练算力。为了在数百个下游领域实现大规模联合强化学习（RL），我们优化了技术栈的每一层——包括 RL 框架、长上下文推理、训练效率、沙箱基础设施、部署调度以及用于环境质量检查的共享平台——从而确保每一份 RL 算力的投入都能带来稳定、可预测的回报。因此，K3.1 Pro 在各个领域都取得了持续的进步。

Across coding, agentic, and knowledge benchmarks, K3.1 Pro matches or approaches the strongest proprietary models. On Terminal-Bench 4.0, which evaluates coding agents on realistic, resource-constrained terminal tasks using strict task-specific tests, K3.1 Pro scores 65.7%, ahead of GPT-6 Astra (58.2%) and narrowly behind Claude Opus 5.5 (66.4%). And on Kimi Code Bench 3.0 Long-horizon, our internal benchmark that tests coding agents’ multi-step execution, tool use, and end-to-end problem solving across diverse engineering domains, K3.1 Pro scores 61.2%, above K3’s 55.95%.

在编程、智能体和知识基准测试中，K3.1 Pro 的表现与最强的闭源模型相当或接近。在 Terminal-Bench 4.0（该测试通过严格的任务特定测试来评估编程智能体在现实、资源受限的终端任务中的表现）中，K3.1 Pro 得分为 65.7%，领先于 GPT-6 Astra（58.2%），略低于 Claude Opus 5.5（66.4%）。而在我们内部用于测试编程智能体在不同工程领域的多步执行、工具使用和端到端问题解决能力的 Kimi Code Bench 3.0 Long-horizon 测试中，K3.1 Pro 得分为 61.2%，高于 K3 的 55.95%。

Coding K3.1 Pro covers the full range of coding work, from quick fixes in real codebases to full-project builds. It also brings a more considered visual style to games and web experiences, and knows how to build them with the right tools. K3.1 Pro works effectively across game engines such as Unity, Unreal Engine, and Godot, combining them with web search and 3D generation to create complete experiences. Its front-end pages, interactive Three.js scenes, and 3D assets show stronger composition, typography, animation, and polish. The examples below were created end to end by K3.1 Pro:

编程。K3.1 Pro 涵盖了从真实代码库中的快速修复到完整项目构建的全方位编程工作。它还为游戏和网页体验带来了更具考究的视觉风格，并懂得如何使用合适的工具进行构建。K3.1 Pro 能在 Unity、Unreal Engine 和 Godot 等游戏引擎中高效工作，并将其与网络搜索和 3D 生成相结合，创造出完整的体验。其前端页面、交互式 Three.js 场景和 3D 资产展现出了更强的构图、排版、动画和精致度。以下示例均由 K3.1 Pro 端到端创建：

K3.1 Pro’s long-horizon coding capabilities are also strengthened through large-scale reinforcement learning. In one internal test, it built a human walking dynamics simulator with integrated training and evaluation pipelines on top of human-atlas, an open-source anatomical model of the human body comprising 296 skeletal components, including cartilage, and 402 muscle components. Across 167 missions and 420 independent reviews, it analyzed experimental results, diagnosed problems, and iteratively refined the walking controller, enabling the simulated human to go from collapsing within seconds to walking continuously with alternating strides for 11.2 seconds.

K3.1 Pro 的长程编程能力也通过大规模强化学习得到了加强。在一项内部测试中，它基于 human-atlas（一个包含 296 个骨骼组件（含软骨）和 402 个肌肉组件的开源人体解剖模型）构建了一个集成训练和评估流程的人体行走动力学模拟器。在 167 项任务和 420 次独立审查中，它分析了实验结果、诊断了问题并迭代优化了行走控制器，使模拟人从几秒内倒下转变为能够以交替步态连续行走 11.2 秒。

When reward shaping failed to improve arm motion, K3.1 Pro did not simply keep adjusting the reward: it ran open-loop probes of the simulator’s dynamics, identified and corrected 36 errors in shoulder-muscle moment-arm calculations, and then retrained the controller.

当奖励塑造无法改善手臂运动时，K3.1 Pro 并没有仅仅不断调整奖励：它对模拟器的动力学进行了开环探测，识别并纠正了肩部肌肉力臂计算中的 36 个错误，然后重新训练了控制器。

3D Modeling and Design K3.1 Pro is strong at 3D creation and spatial reasoning. On BenchCAD, which tests 3D reconstruction from multi-view renderings through CAD code generation, K3.1 Pro scores 92.7% with tools, approaching the performance of Claude Opus 5.5 (96.2%) and GPT-6 Astra (95.9%). From a text brief, sketch, or reference image, it can infer shape, proportions, and spatial relationships to create usable 3D models. It also has a stronger grasp of existing asset libraries and model collections.

3D 建模与设计。K3.1 Pro 在 3D 创作和空间推理方面表现出色。在 BenchCAD（通过 CAD 代码生成测试从多视图渲染中进行 3D 重建）中，K3.1 Pro 在使用工具的情况下得分为 92.7%，接近 Claude Opus 5.5 (96.2%) 和 GPT-6 Astra (95.9%) 的水平。根据文本简报、草图或参考图像，它可以推断形状、比例和空间关系，从而创建可用的 3D 模型。它对现有资产库和模型集合也有更强的掌握能力。

K3.1 Pro particularly excels at parametric modeling, from mechanical parts and architectural models to configurable products and custom furniture. Without MCP connectors, it can write and run scripts for tools like Blender and CAD software to build and refine assets, edit components, and adjust geometry, materials, and scene layout. Below are examples of 3D assets K3.1 Pro created for brand films, games, and interactive web experiences: 3D Make is now live on kimi.ai and the Kimi app — preview and edit your 3D models online, download them ready for 3D printing, or export them as videos or Live Photos to share anywhere.

K3.1 Pro 在参数化建模方面尤为擅长，涵盖从机械零件、建筑模型到可配置产品和定制家具。无需 MCP 连接器，它就能为 Blender 和 CAD 软件等工具编写并运行脚本，以构建和优化资产、编辑组件，并调整几何形状、材质和场景布局。以下是 K3.1 Pro 为品牌电影、游戏和交互式网页体验创建的 3D 资产示例：3D Make 现已在 kimi.ai 和 Kimi App 上线——您可以在线预览和编辑 3D 模型，下载用于 3D 打印的文件，或将其导出为视频或实况照片进行分享。

Finance With stronger reasoning and agentic capability, K3.1 Pro is capable of end-to-end delivery of complete financial workflows. K3.1 Pro achieves state-of-the-art results on OfficeQA Pro v2 (64.4), which tests complex fiscal data analysis over real financial documents through retrieval, cross-document verification, and multi-step numerical reasoning. K3.1 Pro also improves over K3 on Frontier Finance, which measures support for complete investment workflows from screening to profitability analysis.

金融。凭借更强的推理和智能体能力，K3.1 Pro 能够端到端地交付完整的金融工作流。K3.1 Pro 在 OfficeQA Pro v2 (64.4) 上取得了最先进的结果，该测试通过检索、跨文档验证和多步数值推理，评估对真实金融文档的复杂财务数据分析能力。在衡量从筛选到盈利分析的完整投资工作流支持能力的 Frontier Finance 测试中，K3.1 Pro 也较 K3 有所提升。

K3.1 Pro also improves significantly across all our internal finance benchmarks: Financial Modeling evaluates formula linkage, financial logic, and final workbook quality in real-world modeling tasks. K3.1 Pro improves on K3 by 7.2 points. Financial Research evaluates whether models can produce auditable research under evidence and cutoff constraints. K3.1 Pro improves on K3 by 11.6 points. Business Operation evaluates long-horizon management and decision-making in a simulated business environment. K3.1 Pro achieves about 4.2× K3’s performance.

K3.1 Pro 在我们所有的内部金融基准测试中也都有显著提升：金融建模评估真实建模任务中的公式链接、财务逻辑和最终工作簿质量，K3.1 Pro 较 K3 提升了 7.2 分。金融研究评估模型在证据和截止日期限制下能否产出可审计的研究报告，K3.1 Pro 较 K3 提升了 11.6 分。业务运营评估模拟商业环境中的长程管理和决策能力，K3.1 Pro 的表现约为 K3 的 4.2 倍。

Financial analysis and investment research demand a high degree of factual accuracy. To address this, we made targeted improvements to K3.1 Pro’s factual accuracy on real-world tasks and challenging prompts. On Kimi Hallu Bench Hard bench, an internal benchmark designed to stress-test factual accuracy, its claim-level hallucination rate fell from 23.4% for K3 to 8.3%, while the share of responses containing at least one factual error was just 42.4% of its predecessor’s rate. (The evaluation is a stress test for factual accura

金融分析和投资研究需要高度的事实准确性。为此，我们针对 K3.1 Pro 在现实任务和挑战性提示词下的事实准确性进行了针对性改进。在旨在压力测试事实准确性的内部基准测试 Kimi Hallu Bench Hard 中，其声明级幻觉率从 K3 的 23.4% 下降至 8.3%，而包含至少一个事实错误的回复比例仅为前代产品的 42.4%。（该评估是对事实准确性的压力测试）