RelayRouterModels / kimi-k3
kimi-k3 API on RelayRouter
kimi-k3 is a text generation (chat) model from Moonshot AI. On RelayRouter it is served by 1 public group, billed per token from a prepaid balance with no platform fee, through the same base URL as every other model: https://relayrouter.io.
Facts
| Model id | kimi-k3 |
|---|---|
| Provider | Moonshot AI |
| Modality | text generation (chat) |
| Protocols | Anthropic Messages API (/v1/messages) |
| Billing | Per token, prepaid balance, failed requests not billed, no platform fee |
| Payment | Credit or debit card via Stripe |
Groups and rates
| Group | Protocol | Rate |
|---|---|---|
| Kimi K3 | Anthropic Messages API | ¥16.32 (about $2.40) per 1M input tokens, ¥81.6 (about $12.00) per 1M output tokens |
Rates are indicative and settle from your prepaid balance; the rate shown in the console at request time is authoritative. Market reference ¥6.8 per $1. Full table on the models page and pricing.
How to call it
Create a key in the console for a group that lists kimi-k3, then call it with that group's native protocol. The quick start covers all three protocols.
import anthropic
client = anthropic.Anthropic(
api_key="your-relayrouter-key",
base_url="https://relayrouter.io",
)
msg = client.messages.create(
model="kimi-k3",
max_tokens=1024,
messages=[{"role": "user", "content": "Hello!"}],
)
print(msg.content[0].text)Questions
How much does kimi-k3 cost on RelayRouter?
The cheapest public group serving kimi-k3 is "Kimi K3": ¥16.32 (about $2.40) per 1M input tokens, ¥81.6 (about $12.00) per 1M output tokens. Usage is deducted from your prepaid balance and the exact rate at request time is shown in the console.
Which API protocol do I use for kimi-k3?
Use the native protocol of the group you pick: Anthropic Messages API at /v1/messages. The base URL is always https://relayrouter.io and the key goes in the usual header for that SDK.
Are failed requests billed?
No. Only successfully completed calls are billed, and there is no platform fee on top of token usage.