此语言仍有 24 条内容或界面任务待翻译;未完成部分显示中文原文。
AI Home / Model Library / Kimi K3.1 Pro

Kimi K3.1 Pro

AI 辅助整理
月之暗面深度思考函数调用上下文缓存Agentic 编程长程任务Enter Reference Price · Official Prices Prevail

Kimi K3.1 Pro is Moonshot AI's flagship model, launched via the Volcano Ark Agent Plan on 2026-10-24, designed for enterprise-grade long-horizon tasks and Agentic coding. It supports deep thinking, function calling, and context caching, with 5min/1h cache write tiers, making it suitable for complex engineering and agent applications. Note that it is accessed via subscription plans, and the source does not provide figures for context length or pricing, which should be verified in the console.

Context tokensPending Verification
Input / Million tokensPending Verification
Output / Million tokensPending Verification
Versionkimi-k3.1-pro
Deployment通过火山方舟 Agent Plan 套餐调用,可在控制台选择或切换目标模型,也支持在 Claude Code、Codex CLI、OpenCode、TraeCode、Hermes Agent 等工具中配置 Model Name 使用

Capabilities & Use Cases

深度思考函数调用上下文缓存Agentic 编程长程任务

The source is the Volcano Ark "Model Launch Announcement" page (updated 2026-11-20), where the kimi-k3.1-pro entry states: Kimi K3.1 Pro is Moonshot AI's flagship model, designed for enterprise-grade long-horizon tasks and Agentic coding, supporting deep thinking, function calling, and context caching, with 5min/1h cache write tiers, suited to complex engineering and agent applications; the onboarding entry is the "Quick Start" documentation. Deployment and switching: The model is offered through the Agent Plan subscription. If a kimi-latest configuration is used, it currently resolves to kimi-k2.7-code rather than k3.1-pro (the source's latest mapping table contains no kimi-k3.1-pro entry), so to use K3.1 Pro you must write the specific Model Name in the tool configuration file, involving configuration entries for tools such as Claude Code, OpenClaw, Codex CLI, OpenCode, TraeCode, and Hermes Agent. Adjacent versions launched on the same page around the same time include kimi-k2.8-preview (09.17, supporting a 1024K context window) and kimi-k3 (07.17, 1M token context window), but the source does not state whether K3.1 Pro inherits these context window figures, so no inference should be drawn from this. Selection tip: If you need a long-context coding model and must confirm context length, the source only clarifies the context capabilities of kimi-k3 and kimi-k2.7-code, kimi-k2.8-preview; the pricing and quotas of K3.1 Pro are also not given in the source, so verify on the Volcano Ark model plaza and subscription pages before purchasing.

Capability Comparison

No verified item-by-item scores available; capability radar and overall ranking will not be displayed for now.

Data & Sources

Data Provider
月之暗面
On-site profile update
2026-11-22 03:33:16

On-site update time is not equivalent to the model release date. Official or platform benchmark scores are not displayed when test sets, dates, or original results are not provided.

Version History

kimi-k3.1-proCurrent Entry Version

Historical versions, release and discontinuation dates to be verified.

Access & Usage

Please confirm the API address, model identifier, authentication, permissions, and billing rules according to the official documentation.

Open Official Documentation

Platform Tested

No verified model test reports available. Model capability tags do not indicate completed platform testing.
Product Selection Assistant

让需求变成可修改的筛选条件

依据已录入参数生成条件,结果可查看和修改。帧率与性能需核对实测,不能仅凭算力数值保证。