Moonshot AI · Text model
Kimi K3.1 Pro
kimi-k3.1-proModel pricing
USD per one million tokens. Input, output and cache tokens are billed at their own rates.
- Input / 1M tokens
- $1.26
- Output / 1M tokens
- $5.76
- Cache read / 1M tokens
- Not reported
- Cache write / 1M tokens
- Not reported
The final cost depends on actual usage. Model limits below describe the upstream API; Chat may use a smaller output limit.
Estimate a request Input + output
Editable example: 1,000 input and 500 output tokens. Enter token counts from your workload, including any conversation history sent as input. No request will be sent.
- Input
- $0.00126
- Output
- $0.00288
Based on this model’s published catalog rates. Cache reads and writes are excluded. Actual token usage and the channel that serves your request can change the final charge. Model limits do not guarantee this exact input/output combination is accepted.
Choose how to use it
Use the same prepaid account in Chat or connect your own application.
- Max input tokens
- 1,000,000
- Max output tokens
- 32,000
- Image input
- Not supported
- Reasoning
- Supported
- Streaming
- Supported
Your first API request
Use your OpusGate key. Sending this example makes a paid request against your balance.
from openai import OpenAI
client = OpenAI(
api_key="YOUR_OPUSGATE_KEY",
base_url="https://api.opusgate.dev/v1",
)
response = client.chat.completions.create(
model="kimi-k3.1-pro",
messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)Read the connection guides ↗Connection guides
Check your tool’s supported models and custom-provider settings before connecting.
Claude Code setup · Cline setup · OpenCode setup · API reference