Choose a model for the work you need to doMe
Model specifications and sources
QINIU CLOUD官方模型广场对该API region型号的说明：Kimi K3 是 Kimi 迄今能力最强的旗舰模型，拥有 2.8 万亿Parameter，基于 KDA 混合线性注意力机制（Kimi Delta Attention）和注意力残差（Attention Residuals）技术构建，原生支持视觉理解，并拥有 100 万 token Context window。它是Global首个开源的 3 万亿级别模型，面向长程Coding、知识工作和推理等前沿智能场景而设计。
- Context window
- 1,048,576 tokens
- Maximum input
- Not published / unverified
- Maximum output
- Not published / unverified
- Input modalities
- Text, Image, Video
- Output modalities
- Text
- Default snapshot
- Not published / unverified
- Documentation checked
- 2026-09-24
- Automatic model identity check
- 官方本次资料未列出精确型号 ID · 2026-11-12
- Next automatic check
- 2026-11-13T05:21:53.301Z
- Identity check source
- Official evidence page ↗
- Documentation source
- View source ↗
Model recordOfficial documentation未确认Official website标注服务区域：cn
QINIU CLOUD官方模型广场对该API region型号的说明：Kimi K3 是 Kimi 迄今能力最强的旗舰模型，拥有 2.8 万亿Parameter，基于 KDA 混合线性注意力机制（Kimi Delta Attention）和注意力残差（Attention Residuals）技术构建，原生支持视觉理解，并拥有 100 万 token Context window。它是Global首个开源的 3 万亿级别模型，面向长程Coding、知识工作和推理等前沿智能场景而设计。
Writing & chatVision & OCRCoding
Only catalog information is available. The API base URL, request path and authentication have not been verified, so no executable configuration is provided.
API BASE URLNo data available
Model ID
moonshotai/kimi-k3Request method / pathNo data available
AuthenticationNo data available
Visit the provider website or documentation for integration details