Skip to content
Kimi

Kimi K3

deepinfra/moonshotai/Kimi-K3

Kimi K3 is a reasoning model by Kimi with vision, function calling, structured output — designed for complex, multi-step problem solving where accuracy matters more than speed. It features an extended 1M token context window, suitable for processing long documents and complex conversations. Priced at $2.85 per million input tokens and $14.25 per million output tokens. Access Kimi K3 through Inworld LLM Inference with OpenAI SDK compatibility, built-in failover, and intelligent routing across providers.

Input
$2.85
Output
$14.25
Cache read
$0.285
Context
1M
Max output
—
TextImage→Text
Function CallingReasoningStructured outputVision
Run on your server. Set INWORLD_API_KEY to your API key. Python requires pip install openai; JavaScript requires Node.js 18+; save it as example.mjs.
curl --fail-with-body https://api.inworld.ai/v1/chat/completions \
  -H "Authorization: Basic $INWORLD_API_KEY" \
  -H 'Content-Type: application/json' \
  -d '{
    "model": "deepinfra/moonshotai/Kimi-K3",
    "messages": [{"role": "user", "content": "Hello"}],
    "stream": true
  }'
Model Kimi K3OpenAI SDK compatible

Pricing and providers

Access Kimi K3 through Inworld LLM Inference or speech to speech via DeepInfra. By using the providers below you agree to their terms, listed under each provider's name in the Legal section.

Modalities
DeepInfraThis pagedeepinfra/moonshotai/Kimi-K3
$2.85$14.25$0.285—1M—
TextImage→Text
Prices in USD per 1M tokens.

Start building with Kimi K3

Access Kimi K3 and every other model through LLM Inference or speech to speech. Create a free account and start in minutes.