Moonshot AI Models on AtlasCloud | Kimi

Atlas Cloud 通过 MoonshotAI API 提供完整的 Kimi 系列,从用于深度推理的 K2-Thinking 到用于智能体编程的 K2.6。全部按量付费,262K 上下文。

Large Language Models by Moonshot AI

在 Atlas Cloud 上,依托顶尖的大语言模型,以极速且实惠的方式驱动对话、推理与智能体的规模化应用。

查看全部模型

Moonshot AI 模型 API 价格明细

对比 Moonshot AI 全部模型的标准价、平台价与折扣。

模型标准价 (USD)平台价 (USD)折扣
Kimi K2.5
$0.6/$3每 100 万 tokens262.1K 上下文
$0.49/$2.5M 输入/输出每 100 万 tokens262.1K 上下文
查看
Kimi-K2-Thinking
$0.6/$2.5每 100 万 tokens262.1K 上下文
$0.6/$2.5M 输入/输出每 100 万 tokens262.1K 上下文
查看
Kimi-K2-Instruct-0905
$0.6/$2.5每 100 万 tokens262.1K 上下文
$0.6/$2.5M 输入/输出每 100 万 tokens262.1K 上下文
查看
Kimi-K2-Instruct
$0.6/$2.5每 100 万 tokens131.1K 上下文
$0.7/$2.5M 输入/输出每 100 万 tokens131.1K 上下文
查看

探索其他厂商的模型

在 Atlas Playground 中即刻探索并体验 300+ 生产可用模型,一键开始定制。

您可以在 Atlas Cloud 上构建的 Moonshot AI API 应用案例

Kimi 的智能体集群和长周期执行能力,使团队能够在单次自动化会话中完成原本需要数天人力才能完成的任务。团队将 M-series 与 K2-Thinking 结合使用,从而涵盖从自主代码更改到大规模多文档研究的各项工作。

遗留代码库现代化

工程团队使用 Kimi K2.6 运行长周期编码智能体,在长达数小时的会话中自主重构生产环境的代码库。在一个有记录的案例中,K2.6 耗时13个小时重写了一个有着8年历史的金融撮合引擎,在各次提交之间无需人工干预的情况下,将吞吐量提升了185%。Atlas Cloud 的按需付费定价模式使得运行这些长时间的智能体任务变得切实可行,且无需预先承诺容量。

并行文档批量处理

运营团队使用 Kimi K2.6 的 300 智能体集群并行处理大批量文档。单次编排运行可将一份简历与 100 个岗位角色进行匹配,并输出 100 份完全定制的简历。相同的模式适用于合同审查、合规检查,以及任何需要将固定输入与大量、可变的目标集进行评估的工作流。

面向复杂分析的深度推理

研究和法律团队使用 Kimi K2-Thinking 处理需要长时间内部推理的多步分析问题。该模型每次会话支持多达 200 到 300 次连续的工具调用,在“推理-调用-推理”的循环中迭代,步骤之间无需人工提示。在 Atlas Cloud 上,其定价为每百万输入 token 0.6 美元,并与 Kimi 系列的其他产品共享 262K 的上下文窗口。

自动化研究论文生成

学术和内容团队使用 Kimi K2.6 将源文档转化为完整的研究成果。在一次演示运行中,K2.6 在单次会话内将一篇天体物理学论文转换成了一篇40页的研究论文、一个包含超过20,000个条目的结构化数据集以及14张专业天文图表。这将从文献到成果的工作流周期从数周缩短至数小时。

规模化潜在客户挖掘

增长和销售团队使用 Kimi K2.6 集群并行识别潜在客户并生成外拓素材。在一个运行示例中,它在一个目标城市识别出30家没有网站的零售店,并为每家店生成了一个落地页。这种模式同样适用于线索丰富、竞争格局梳理,以及任何在大批量列表上结合了发现与内容生成的任务。

视觉文档与代码分析

产品和数据团队使用 Kimi K2.5 和 K2.6 的原生视觉能力,在同一次 API 调用中处理图像和视频输入以及文本。MoonViT 编码器无需外部预处理即可处理图表、屏幕截图、UI 界面原型和文档扫描件。这对于将视觉规范直接转化为代码,或从包含大量图像的文档中提取结构化数据的流水线非常有用。

Atlas Cloud AI

联系销售

关于 Moonshot AI 模型的常见问题解答

Kimi K2.6 is MoonshotAI's latest open-source multimodal LLM, released in April 2026 under a Modified MIT license. It runs a Mixture-of-Experts architecture with 1 trillion total parameters and 32 billion active during inference. It is designed for agentic coding, long-horizon task execution, and multi-agent swarm orchestration.

Kimi K2.6 可扩展至300个子智能体,在单次运行中执行多达4000个协同步骤。Atlas Cloud 上的 Kimi K2.5 支持多达100个子智能体的集群执行。任务被动态分解为并行的、领域专用的子任务,以实现完全自主的输出。

Kimi K2-Thinking 采用深度思维链推理,每次会话最多可进行 200 到 300 次连续的工具调用。该模型进行推理、调用工具、解释结果、调用另一个工具,并在无需人工输入的情况下继续这一循环。它适用于多步逻辑推演、复杂数学问题以及通过扩展内部推理来提高准确性的问题。

是的。Kimi K2.5 和 K2.6 包含了 MoonViT,这是一个拥有 4 亿参数的视觉编码器,能够原生处理图像和视频。您可以在 API 调用中直接将图像或视频输入与文本一起传递,无需进行外部预处理。这支持视觉分析、文档理解以及图像到代码生成的工作流。

是的。Kimi K2.6 基于修改版的 MIT 许可证发布,允许商业使用。开源权重可在 HuggingFace 上获取,用于自托管部署。对于倾向于托管访问而无需承担基础设施开销的团队,Atlas Cloud 也通过 API 提供 K2.6。

Kimi K2.6在SWE-Bench Verified上得分为80.2%,在使用工具的Humanity's Last Exam中得分为54.0%,在这两项基准测试中均优于GPT-5.5。它在BrowseComp上也以83.2%的成绩领先,高于GPT-5.4。取得这些结果的每百万token成本比GPT-5.5低约80%。

Kimi K2.5 在 Atlas Cloud 上的定价为每百万输入 tokens 0.49 美元,每百万输出 tokens 2.5 美元。Kimi K2-Thinking 和 K2-Instruct-0905 的运行价格为每百万输入 tokens 0.6 美元,输出费率相同。请查看 Atlas Cloud 的 Kimi K2.6 模型页面以获取其当前具体的定价信息。

探索更多系列

Seedance 2.0 Models

Seedance 2.0(by Bytedance) is a multimodal video generation model that redefines "controllable creation," moving beyond the limitations of text or start/end frames. It supports quad-modal inputs—text, image, video, and audio—and introduces an industry-leading "Universal Reference" system. By precisely replicating the composition, camera movement, and character actions from reference assets, Seedance 2.0 solves critical issues with character consistency and physical coherence, empowering creators to act as true "directors" with deep control over their output.

查看系列

Grok-Imagine Models

Grok Imagine Image Quality is xAI's latest AI image generation model, delivering studio-grade visuals with up to 2K resolution and razor-sharp detail. It offers best-in-class text rendering across multiple languages, photorealistic outputs with natural lighting, rich textures, and believable physics, plus tighter prompt following and image editing with reference inputs for precise creative control. Ideal for hero images, ad creatives, product renders, and brand-grade visuals.

查看系列

Gemini Omni

Gemini Omni (by Google DeepMind) is a video generation and editing model launched on May 20, 2026 at Google I/O that redefines the standard for "reasoning-driven creation," built specifically to solve the core challenge of AI video: making output that actually understands what you mean, not just what you type. It fuses Gemini's reasoning engine with generative capability, accepting any mix of images, text, video, and audio to produce consistent, knowledge-grounded output. Unlike models that start from scratch each time, Omni lets you edit through natural conversation — swapping objects, rewriting scenes, shifting styles — while keeping physics, characters, and continuity intact across every turn.

查看系列

GPT Image 2 Models

GPT Image 2 is a state-of-the-art multimodal foundation model engineered for exceptional text-to-image generation with unprecedented photorealism and creative versatility. Developed by OpenAI as the evolution of the DALL-E lineage, it transforms detailed natural language descriptions into hyper-realistic imagery at up to 4K resolution. With proprietary "Neural Rendering Engine" technology for precise visual control, GPT Image 2 delivers studio-quality results with accurate anatomy, lighting, and composition—making it the premier AI tool for professional creators, enterprises, and developers demanding production-ready visual assets.

查看系列

Google Models on Atlas Cloud | Gemini, Nano Bananas & Veo

Google最强大的创意模型现已在Atlas Cloud上全面可用。Veo 3.1提供电影级别的视频生成,Nano Banana 2支持高保真图像创建,而Gemini为每个工作流带来多模态智能。通过单一API key即可访问完整的Google模型套件,提供Day-0可用性和按需付费(pay-as-you-go)定价。

查看系列

ByteDance Models on Atlas Cloud | Seedance & Seedream

从电影级视频生成到高保真图像创建,ByteDance 最强大的模型现已在 Atlas Cloud 上线。以最低的推理定价和零基础设施开销,大规模运行 Seedance 和 Seedream。

查看系列

Alibaba Models on Atlas Cloud | Wan & Qwen

Atlas Cloud 将 Alibaba 的全系模型阵容整合至同一个 API 中:Qwen 用于语言和图像任务,Wan 用于高达 1080p 的视频生成。所有模型均采用按需付费模式,无需订阅。您可以使用现有的 OpenAI 兼容客户端,通过单一的 base URL 访问 Alibaba API。

查看系列

MAI Image 2.5 Models

MAI-Image-2.5 是 Microsoft 最新推出的逼真图像生成与编辑模型系列,专为商业设计、产品摄影和品牌级内容创作而打造。提供用于文本生成图像和图像编辑的 standard 和 Flash 变体,以极具竞争力的价格(每张图像起价 0.03 美元)提供同类最佳的 Arena ELO 得分。凭借精准的文本渲染、手术刀级的编辑能力以及自然的人像生成,MAI-Image-2.5 专为需要生产级质量视觉效果且无需承担后期处理开销的团队而设计。

查看系列

Wan2.7 Models

Launching this March, Wan2.7 is the latest powerhouse in the Qwen ecosystem, delivering a massive upgrade in visual fidelity, audio synchronization, and motion consistency over version 2.6. This all-in-one AI video generator supports advanced features like first-and-last frame control, 3x3 grid synthesis, and instruction-based video editing. Outperforming competitors like Jimeng, Wan2.7 offers superior flexibility with support for real-person image inputs, up to five video references, and 1080P high-definition outputs spanning 2 to 15 seconds, making it the premier choice for professional digital storytelling and high-end content marketing.

查看系列

Nano Banana2 Models

Nano Banana 2 (by Google), is a generative image model that perfectly balances lightning-fast rendering with exceptional visual quality. With an improved price-performance ratio, it achieves breakthrough micro-detail depiction, accurate native text rendering, and complex physical structure reconstruction. It serves as a highly efficient, commercial-grade visual production tool for developers, marketing teams, and content creators.

查看系列

Midjourney Models

Midjourney is a proprietary AI image and video generation platform developed by Midjourney, Inc. (San Francisco). Founded in 2021 by David Holz, it has become the aesthetic gold standard in generative AI — transforming text prompts into cinematic, painterly visuals at native 2K resolution. The latest V8.1 architecture, rebuilt from scratch on GPU-native PyTorch, delivers 4–5× faster generation, true 2048×2048 output without upscaling artifacts, and a signature visual style that remains unmatched by competitors. With the addition of Video V1, Midjourney extends its aesthetic into motion — animating still images into atmospheric 5-second cinematic clips. From brand campaigns to film pre-visualization to game concept art, Midjourney is the premier AI creative tool for professionals who demand both speed and artistry.

查看系列

PixVerse Models

PixVerse, developed by AISphere, is a video generation model series built around one idea: giving creators director-level control over every frame. V6 is the flagship generation model, covering text-to-video, image-to-video, reference-to-video, start-and-end frame control, and video extension in a single cohesive pipeline. C1 takes a different approach — it is a storyboard-native model designed for multi-shot narrative production, where scene continuity and visual consistency across clips matter as much as individual frame quality. Both series are available on Atlas Cloud, starting from $0.025 per second, with no infrastructure setup required.

查看系列

推荐文章

指南、教程与产品动态,助你充分发挥 Atlas Cloud 的价值。

Join our Discord community

Join the Discord community for the latest model updates, prompts, and support.