引入 Kimi K3.1 Pro
我们介绍Kimi K3.1 Pro，我们新的K3.1系列中的第一个模型，以及我们迄今为止最有能力的3T级开放量级模型，有一个100万托肯的上下文窗口。 它在编码、三维创建、金融研究和其他代理能力方面都有重大改进。
我们还在释放Kimi K3.1 Pro高速，它能提供K3.1 Pro的2-3×速度，以进行更灵敏的互动。
Kimi K3.1 Pro和Kimi K3.1 Pro高速今日可在Kimi.ai,Kimi Work,Kimi代码和Kimi API上获得。 到11月4日，完整的型号重量将是发布。
推进开放情报前沿
Kimi K3.1 号机车 Pro是Kimi K3的训练后版本，训练后计算量要大得多。 为了能够跨过数百个下游域进行大规模联合RL，我们优化了堆积层的每一层——RL框架，长文推论，培训效率，沙盒基础设施，推出时间表，以及环境质量检查共享平台——这样，RL计算的每一增量都能够提供稳定，可预测的回报。 因此，K3.1 Pro在每个领域都取得了一致的收益。
通过编码、代理和知识基准测试，K3.1 Pro匹配或接近最强的专利模型。 在使用严格的任务特定测试对agents进行现实性，资源约束的终端任务进行编码的终端-Bench 4.0上，K3.1 Pro得分65.7%，领先于GPT-6 Astra(58.2%)，窄于Claude Opus 5.5(66.4%). 在Kimi代码3.0长视线上，我们的内部基准测试测试编码agents的多步执行，工具使用，以及端到端的问题解决跨越不同的工程领域，K3.1 Pro得分61.2%，高于K3的55.95%.
编码
K3.1 Pro涵盖从真实代码库中的快速解决方案到全项目建设的全部编码工作。 同时也是带来对于游戏和网络体验的更深思熟虑的视觉风格，并且知道如何用正确的工具来构建它们。
K3.1 Pro在Unity,unreal Engine，和Godot等游戏引擎之间有效工作，结合了网页搜索和3D生成来创造完整的体验。 其前端页面，交互式Tree.js场景，以及3D资产显示更强的构成，打字，动画，和抛光。 以下例子由 K3.1 Pro 创建为结束：
K3.1 Pro的长视码编码能力也通过大规模强化学习得到加强。 在一次内部测试中，它搭建了人类行走动力模拟器，在人-阿特拉斯顶部设有集成的训练和评价管道，人体的开源解剖模型包括包括软骨在内的296个骨骼元件和402个肌肉元件。 在167次任务和420次独立审查中，它分析了实验结果，诊断出问题，并迭代地改进了行走控制器，使模拟人类能够从几秒钟内崩溃到连续走11.2秒的交替行走。 当奖励塑造未能改进手臂运动时，K3.1 Pro并没有简单地继续调整奖励：它运行了模拟器动力学的开球探测器，识别并纠正了肩部肌肉瞬间臂计算中的36个错误，然后重新训练控制器。
3D 建模和设计
K3.1 Pro在3D创建和空间推理上很强。 在通过CAD代码生成测试多视图渲染的3D重建的BeningCAD上，K3.1 Pro用工具评分92.7%，接近Claude Opus 5.5(96.2%)和GPT-6 Astra(95.9%)的性能。 从文本简报，素描，或参考图像中，它可以推断形状，比例，和空间关系来创建可用的3D模型。 它还更能掌握现有资产图书馆和示范收藏。
K3.1 Pro在从机械部件和建筑模型到可配置产品和定制家具的参数模型制作方面特别出色。 没有MCP连接器，它可以为Blender和CAD软件等工具编写和运行脚本，以构建和完善资产，编辑组件，并调整几何，材料，场景布局。
下面是三维资产的例子 K3.1 为品牌电影、游戏和互动网络体验创建的Pro:
3D Make现在在kimi.ai和Kimiapp上直播——在线预览和编辑你的3D模型，下载这些模型准备进行3D打印，或者导出为视频或Live Photos以分享任何地方。
财务
有了更强的推理和代理能力，K3.1 Pro能够端到端交付完整的财务工作流程。 K3.1 Pro在OfficeQA Pro v2(64.4)上实现了最先进结果，该结果通过检索，跨文件验证，以及多步骤的数字推理，对真实财务文件进行复杂的财政数据分析进行测试。 第3.1条 Pro还改进了 " K3 " 的 " 边境金融 "，该 " K3 " 衡量对从筛选到利润分析的完整投资流程的支持。
K3.1 Pro也大大改善了我们所有的内部财务基准测试：
- 金融建模在现实世界建模任务中评价公式链接、金融逻辑和最后工作手册质量。 K3.1 Pro在K3上改进了7.2分。
- 金融研究评价模型是否能够在证据和截断限制下产生可审计的研究。 K3.1 Pro在K3上提高了11.6分。
- 业务业务评价模拟业务环境中的长期管理和决策。 第3.1条 Pro取得了大约4.2×K3的性能。
财务分析和投资研究需要高度的事实准确性。 为了解决这一问题，我们有针对性地改进了K3.1 Pro在现实世界任务和具有挑战性的提示方面的事实准确性。 在Kimi Hallu Bench Hard 板凳上，一个内部的基准测试设计用来测试事实准确性，其主张水平的幻觉率从K3的23.4%下降到8.3%，而至少包含一个事实错误的回复比例仅为其前身的42.4%. (评价是对事实准确性的一种压力测试，其中的一些实例是有意用来调查模型容易出错的情况。) 报告的比率并不代表每日使用的总误差率。 )
以下案例显示K3.1 Pro如何在实践中处理金融模型、投资分析和来源研究：
网络安全
作为 K3.1 Pro变得更有能力了，我们花费了相应的更大份额的努力来理解网络能力来自哪里，以及我们能控制多少。 因此，我们分析和评估了K3.1 Pro的脆弱性研究能力，并在培训期间采用有针对性的控制措施，以减少潜在的滥用风险，同时保持总体效用。
识别并删除网络相关数据。 在开发过程中，我们看到随着培训后规模的扩大，脆弱性研究的表现稳步改善。 追踪这一趋势的来源，我们发现，与网络有关的内容植根于一般的中期培训中，与网络有关的任务存在于RL环境组合中——这是广泛数据培训的自然副产品。 在对内部模型的初步实验中，我们发现，如果不在中培训阶段或RL阶段删除数据，下游的网络能力将大大提高。 这些发现给了我们具体的控制杠杆。 对于K3.1 Pro的开放量发布，我们从两个培训阶段都识别并移除了网络相关数据，同时保留了模型实用性所需的通用编码任务。 我们将继续在每个培训阶段实施适当的控制，其中与网络有关的数据识别和限制就是一个例子。
评价和结果。 在数据清理之后，我们评估了Xdev-bench-hard上的K3.1 Pro，我们内部的基准测试基于专家验证的源码脆弱性发现和开发任务，校准了这些任务，以便在我们目前的能力水平上提供有意义的区分。 大多数任务在最近忠实地复制了Linux内核，Node.js,JavaScriptCore,Exim，和VirGL等软件中披露的弱点，并启用了缓解功能。 每个任务都提供了脆弱的源代码，一个短的bug提示，以及相应的二进制；agent必须开发一个开发，在单独的沙盒中实现任意的代码执行，并检索一个用于自动化验证的秘密字符串。 我们报告通过@ 3 随机子集的运行，没有困难的超时或输出符号限制。
在xdev-bench-hard上，没有这些降低风险措施的检查站——在中训时没有删除网络相关数据，也没有进行网络相关RL环境过滤——在通过@3时解决90.5%的问题；最后的K3.1 Pro达到81.0%. 数据删除的显示的比较，比起同一培训管道原本会产生的网络能力，有效地降低了网络能力。 与此同时，K3.1 Pro比其前身有了很大的改进，表明即使目标明确的数据删除之后，更强大的一般编码和长视距调试能力也会被普遍应用于网络任务。 第3.1条 Pro的网络能力仍然落后于已经公开至少三个月的前沿专有模式。 我们将不断发展我们的内部基准测试，以便随着示范能力的持续发展，评价能够保持信息性。
开放权重原则。 这些评估为我们开放量释放决定提供了框架。 首先，我们的目标是不释放开放权重，这将扩大网络能力的绝对前沿。 K3.1 Pro的进攻性网络能力，从设计上讲，低于我们的管道所能生产的能力，并且仍然落后于前沿专有模式。 第二，我们打算逐步提高开放权重的网络能力。 我们认为，以非歧视方式提供K3.1方案有利于需要双重用途能力的人。 当一个有能力的模型打开并运行在其用户控制的基础设施上时，更多的团队获得了审计其维持的代码的手段——包括很少接受专门安全审查的遗留系统。 我们认为，让这种能力逐渐扩散——而不是将其集中在少数专有系统——最终使所有软件更加安全。
可用性
Kimi K3.1 号机车 Pro 推出今天横跨Kimi产品家族：在Kimi.ai和Kimi移动应用，Kimi Work和Kimi代码上，并通过Kimi API.
到2026年11月4日，全型号重量将达到发布。
Kimi 企业管理员可以启用 K3.1 Pro for their organization；默认在发布关闭访问。
每个频道的细节：
- Kimi Agents：从您的移动应用商店下载或更新最新的Kimi应用，可在iOS,Android和HarmonyOS上获得，或访问kimi.ai.
- 与 Kimi K3.1 Pro 合作：下载最新的 Kimi 工作桌面应用，版本为 3.1.0 或以后，可供 Windows 和 Apple 硅Macs使用。
- 代码与 Kimi K3.1 Pro：运行 Kimi 代码并选择 Kimi K3.1 Pro使用/型号命令。
- 使用Kimi API构建：访问Kimi API平台并选择kimi-k3.1-pro / kimi-k3.1-pro-高速。 0数据保留为API 客户提供。
——带Kimi到你的组织：Kimi Business提供企业级数据隐私和成员管理，个人账户和组织账户完全分离。 访问定价页面并选择“ Get Kimi Business” 为您的团队订阅。
兹基准测试
脚注
除非另有说明，所有 Kimi K3.1 下文报告的结果是用“最大”、温度=1.0和顶端-p=0.95的推理方法得出的。
编码基准测试
- 4.0. Kimi K3.1终点站 Pro用Claude-Code带评价。 DeepSeek-V4.1-Flash-1017取自发布博客(https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017). 所有剩余的分数都作为博客文章或官方终端-Bench 4.0领导板(https://www.tbench.ai/)中报告的最大值。
- 终端-奔驰科学0.1. Kimi K3.1 Pro用 Claude 代码评价
本文由 AI图鉴自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。
Introducing Kimi K3.1 Pro
We'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.
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.
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%.
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:
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. 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.
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.
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.
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.
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.
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 accuracy, with examples deliberately crafted to probe situations where models are prone to errors. The reported rates do not represent overall error rates in daily use.)
The cases below show how K3.1 Pro handles financial modeling, investment analysis, and source-grounded research in practice:
Cybersecurity
As K3.1 Pro became more capable, we spent a correspondingly larger share of our effort understanding where cyber capability comes from, and how much of it we can control. We therefore analyzed and evaluated K3.1 Pro's vulnerability research capabilities, and applied targeted controls during training to reduce potential misuse risk while preserving general usefulness.
Identifying and removing cyber-related data. During development, we observed vulnerability research performance improving steadily as post-training scaled. Tracing the source of this trend, we found that cyber-related content was embedded within the general mid-training corpus, and cyber-related tasks were present in the RL environment mix — a natural byproduct of training on broad data. In preliminary experiments on internal models, we found that without data removal at either the mid-training or the RL stage, downstream cyber capability would increase substantially. These findings gave us a concrete lever for control. For the open-weight release of K3.1 Pro, we identified and removed cyber-related data from both training stages, while retaining the general-purpose coding tasks needed for model usefulness. We will continue to apply appropriate controls at every training stage, of which cyber-related data identification and restriction is one example.
Evaluation and results. After data cleaning, we evaluated K3.1 Pro on xdev-bench-hard, our in-house benchmark of expert-curated, source-code-based vulnerability discovery and exploitation tasks, calibrated to provide meaningful differentiation at our current capability level. Most tasks faithfully reproduce recently disclosed vulnerabilities in software such as the Linux kernel, Node.js, JavaScriptCore, Exim, and VirGL, with mitigations enabled. Each task provides the vulnerable source code, a short bug hint, and the corresponding binary; the agent must develop an exploit, achieve arbitrary code execution in a separate sandbox, and retrieve a secret string for automated verification. We report pass@3 over a random subset of runs, with no hard timeouts or output token limits.
On xdev-bench-hard, the checkpoint without these risk-reduction measures — no cyber-related data removal at mid-training and no cyber-related RL environment filtering — solves 90.5% of problems at pass@3; the final K3.1 Pro reaches 81.0%. The comparison shows that data removal meaningfully reduces cyber capability relative to what the same training pipeline would otherwise produce. At the same time, K3.1 Pro improves substantially over its predecessor, indicating that stronger general coding and long-horizon debugging capabilities generalize to cyber tasks even after targeted data removal. K3.1 Pro's cyber capability remains behind that of frontier proprietary models that have been publicly available for at least three months. We will keep evolving our internal benchmarks so that evaluation stays informative as model capabilities continue to advance.
Open-weight principles. These evaluations frame our open-weight release decision. First, we aim not to release open weights that would extend the absolute frontier of cyber capability. K3.1 Pro's offensive cyber capability is, by design, below what our pipeline can produce, and remains behind that of frontier proprietary models. Second, we intend to progressively raise the cyber capability available in open weights. We believe making K3.1 Pro available in a non-discriminatory way benefits people who need dual-use capabilities for defensive purposes. When a capable model is open and runs on infrastructure its users control, many more teams gain the means to audit the code they maintain — including legacy systems that rarely receive dedicated security reviews. We believe that letting this capability diffuse gradually — rather than concentrating it in a few proprietary systems — ultimately makes all software more secure.
Availability
Kimi K3.1 Pro rolls out today across the Kimi product family: on Kimi.ai and the Kimi mobile apps, Kimi Work and Kimi Code, and through the Kimi API.
Full model weights will be released by Nov 4, 2026.
Kimi Enterprise administrators can enable K3.1 Pro for their organization; access is off by default at launch.
Details for each channel:
- Kimi Agents: Download or update to the latest Kimi app from your mobile app store, available on iOS, Android, and HarmonyOS, or visit kimi.ai.
- Work with Kimi K3.1 Pro: Download the latest Kimi Work desktop app, version 3.1.0 or later, available for Windows and Apple silicon Macs.
- Code with Kimi K3.1 Pro: Run Kimi Code and select Kimi K3.1 Pro using the /model command.
- Build with the Kimi API: Visit the Kimi API Platform and select kimi-k3.1-pro / kimi-k3.1-pro-highspeed. Zero Data Retention is available for API customers.
- Bring Kimi to your organization: Kimi Business provides enterprise-grade data privacy and member management, with complete separation between personal and organization accounts. Visit the pricing page and select “Get Kimi Business” to subscribe for your team.
Benchmarks
Footnotes
Unless otherwise specified, all Kimi K3.1 Pro results reported below are obtained with the reasoning effort set to 'max', temperature = 1.0, and top-p = 0.95.
Coding benchmarks
- Terminal-Bench 4.0. Kimi K3.1 Pro is evaluated with the Claude-Code harness. The DeepSeek-V4.1-Flash-1017 is taken from the release blog (https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017). All remaining scores are taken as the maximum value reported in either their blog posts or the official Terminal-Bench 4.0 leaderboard (https://www.tbench.ai/).
- Terminal-Bench Science 0.1. Kimi K3.1 Pro is evaluated with the Claude-Code h
Auto-fetched and translated by AI Atlas. All rights belong to the original author.