Kimi K2 Thinking API
Kimi K2 that reasons before it answers: maths, logic, planning.
Good for
Moonshot AI’s models for long tasks and agents; the Thinking version reasons before it answers.
| Model ID | kimi-k2-thinking |
| Vendor | Moonshot AI |
| Family | Kimi |
| Balance rate | ×0.3 / ×1.25 |
| Cache read | $0.015 per 1M tokens |
Both API formats
The model answers in the Anthropic Messages format and in the OpenAI Chat Completions format. Only the address and the key change.
curl https://api.klinch.cloud/v1/messages \
-H "x-api-key: kl_live_…" \
-H "content-type: application/json" \
-d '{"model": "kimi-k2-thinking", "max_tokens": 1024,
"messages": [{"role": "user", "content": "Hello"}]}'from openai import OpenAI
client = OpenAI(base_url="https://api.klinch.cloud/v1", api_key="kl_live_…")
r = client.chat.completions.create(
model="kimi-k2-thinking",
messages=[{"role": "user", "content": "Hello"}],
)
print(r.choices[0].message.content)FAQ
How much does Kimi K2 Thinking cost on KLINCH?
$0.06 per 1M input tokens and $0.25 per 1M output tokens. Moonshot AI’s list price is $0.6 and $2.5. The balance is charged ×0.3 per input and ×1.25 per output token.
How do I call Kimi K2 Thinking through the API?
Use the KLINCH API address, your kl_live_… key and the model kimi-k2-thinking. The Anthropic SDK, the OpenAI SDK, Claude Code, Cursor and any client with a custom base URL work.
Is there a subscription?
No. You buy a token pack once and tokens are charged as you use them.
Is this the official Moonshot AI API?
No. KLINCH is not affiliated with Moonshot AI. We provide access to the model through our own gateway that speaks the Anthropic and OpenAI APIs.
One key, one balance, every top model
Claude, GPT, Gemini, DeepSeek and Grok at a tenth of the official price. No subscription, pay as you go.