kimi-k3
Moonshot AI · Text
Moonshot Kimi K3 is Kimi's most capable flagship yet — a 2.8-trillion-parameter model with native vision and a 1M-token context, built for long-horizon coding, agentic and reasoning tasks. Served on getmodel over an OpenAI-compatible endpoint.
Pricing
| Item | Price |
|---|---|
| Input | $3.135 / 1M tokens |
| Output | $15.675 / 1M tokens |
How to call
curl https://getmodel.ai/v1/chat/completions \
-H "Authorization: Bearer $GETMODEL_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"kimi-k3","messages":[{"role":"user","content":"Hello"}]}' from openai import OpenAI
client = OpenAI(base_url="https://getmodel.ai/v1", api_key="$GETMODEL_API_KEY")
resp = client.chat.completions.create(
model="kimi-k3",
messages=[{"role": "user", "content": "Hello"}],
)
print(resp.choices[0].message.content) Details
FAQ
How much does the kimi-k3 API cost?
kimi-k3 on GetModel costs $3.135 / 1M tokens (Input), $15.675 (Output), billed pay-as-you-go with no subscription.
How do I call kimi-k3 via API?
Point any OpenAI-compatible client at https://getmodel.ai/v1 with your GetModel API key and set the model to "kimi-k3".
Can I use kimi-k3 together with other models?
Yes — one GetModel API key works for kimi-k3 and 30+ other models. Just change the model field.