Kimi K3
Call the default-enabled Kimi K3 model through OpenAI Chat Completions or the Responses API.
Kimi K3 is a default-enabled chat model for general Q&A, content generation, and multi-turn conversations. Treat the live
GET /v1/modelsresponse as the source of truth for availability.
Model information
| Item | Value |
|---|---|
| Model ID | kimi-k3 (when returned by the current API key's model list) |
| Primary endpoint | POST /v1/chat/completions |
| Request format | OpenAI Chat Completions |
| Responses | Supported when the model entry includes responses_endpoint |
| Pricing | Follows the live ClawFarm catalog and your account coefficient |
Use the bare kimi-k3 ID returned by the model list in new integrations. It follows the channel
selected for that model on the current API key.
Request parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
model | string | Yes | Use an ID returned by the current API key's model list, such as kimi-k3 |
messages | array | Yes | Ordered messages containing role and content |
max_tokens | integer | No | Maximum number of tokens in the response |
temperature | number | No | Sampling temperature; the supported range depends on the current upstream |
stream | boolean | No | Set to true to receive an SSE stream |
Request example
curl https://hopsapi.com/v1/chat/completions \
-H "Authorization: Bearer sk-YOUR_KEY" \
-H "content-type: application/json" \
-d '{"model":"kimi-k3","max_tokens":256,
"messages":[{"role":"user","content":"Summarize three benefits of remote work."}]}'Response example
This example shows the protocol shape only. Text, token counts, and additional fields such as id and model come from the live response.
{
"choices": [
{
"message": { "role": "assistant", "content": "1. Greater location flexibility..." },
"finish_reason": "stop"
}
],
"usage": { "prompt_tokens": 18, "completion_tokens": 64, "total_tokens": 82 }
}The response preserves the native OpenAI structure. See OpenAI format and Responses API for streaming and Responses examples.