Text model API
Kimi K3 API
moonshotai/kimi-k3 is Moonshot AI's flagship multimodal reasoning model, built for advanced coding, complex knowledge work, and long-horizon AI agents. With a 2.8 trillion-parameter Mixture-of-Experts (MoE) architecture and a massive 1 million-token context window, Kimi K3 excels at understanding large codebases, analyzing lengthy documents, processing images, and solving multi-step reasoning tasks. Its architecture incorporates Kimi Delta Attention (KDA) and Attention Residuals for highly efficient long-context inference. Designed for developers, researchers, enterprises, and AI-powered appl
- Model id
kimi-k3- Takes
- Text (required)
- Makes
- Text
- Endpoint
POST /api/v1/chat/completions- Made by
- Moonshot AI
- Mode
- Synchronous
Call Kimi K3
https://payload.rentprompts.com/api/v1. Replace the placeholder with your key.Model kimi-k3
POST /api/v1/chat/completions
curl https://payload.rentprompts.com/api/v1/chat/completions \
-H "Authorization: Bearer rp_live_YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "kimi-k3",
"messages": [
{
"role": "user",
"content": "Write a haiku about APIs."
}
],
"max_tokens": 256
}'Replace rp_live_YOUR_KEY with your key, ideally via the RENTPROMPTS_API_KEY environment variable. Keys are shown only once, so they are never filled in here. Get your API key →
What a typical call costs
- 1,000 input + 500 output tokens1.35 Joules
- 10,000 input + 2,000 output tokens8.4 Joules
Send requests to POST /api/v1/chat/completions. Kimi K3 answers synchronously: the result comes back in the response to the same request. Token counts come from the model's own usage report, so a real call may use more or fewer tokens than this example.
Settings
This model has no extra settings. Send the standard fields shown in the example.