Skip to main content

kimi/kimi-k3.1-pro

kimi/kimi-k3.1-pro is a Kimi model whose inference service is provided by Moonshot AI. It supports text and image input and has a context length of 1,048,576 tokens.

Inference Service Provider

The inference service provider for kimi/kimi-k3.1-pro is Moonshot AI.

Model Capabilities

CapabilitySupportCapabilitySupport
Input ModalityText ImageOutput ModalityText
Model ExperienceUnsupportedFunction CallingSupported
Structured OutputsSupportedWeb SearchUnsupported
Prefix CompletionSupportedContext CachingSupported
Batch InferenceUnsupportedFine-tuningUnsupported

Context Limits

ParameterValueParameterValue
Max Input Length1048576Max Output Length1048576
Context Window1048576Max Input Length (Thinking Mode)1048576
Max Output Length (Thinking Mode)1048576Max Chain-of-Thought Length1048576

Pricing

This page only shows the original pricing for model API calls, excluding any limited-time promotions. Visit Model Studio Console for promotional offers.
  • China (Beijing)
Billing ItemPrice (CNY)Unit
Input13Per 1M tokens
Output68Per 1M tokens
Input (Cache Hit)1.3Per 1M tokens

Rate Limits

  • China (Beijing)
ParameterValue
RPM (Requests Per Minute)500
TPM (Tokens Per Minute)3,000,000

Usage limitations

  • The parallel_tool_calls parameter is not supported.
  • The n parameter is not supported.

Invocation

Call the model using the OpenAI-compatible API. For instructions, see Kimi-Moonshot AI.