Kimi K3

Kimi K3 supports long-document analysis, coding, visual understanding and tool workflows.

moonshotai/kimi-k3

Chat with this modelSign in and use your USD balance. No API key required.

Prices & model limits

Input / 1M tokens: $0.16875
Output / 1M tokens: $5.0625
Cache read / 1M tokens: $0.032063
Context window: 1M
Input per request: 1M
Output per request: 943.7K
Protocols: Chat Completions

Limits apply per request. The context window includes both input and output tokens.

Compare market prices

Run the numbers.

Total input tokensTotal output tokens

Estimated total$1.1813

Standard token estimate, before cache or any separately documented pricing tiers.

Suggested workflows

Long-document analysis
Coding and visual tool workflows

Before you integrate

FIRST REQUEST

Ready for your code.

Create an API key

Python

from openai import OpenAI
import os

client = OpenAI(
    base_url="https://lexelane.com/v1",
    api_key=os.environ["LEXELANE_API_KEY"],
)

response = client.chat.completions.create(
    model="moonshotai/kimi-k3",
    messages=[{"role": "user", "content": "Hello, Lexelane."}],
    max_tokens=256,
)
print(response.choices[0].message.content)

FAQ

Good questions. Clear answers.

How should I compare prices?+

Compare input and output rates for the same model version using your own token mix. Caching, context length, service tiers, and quantization can affect comparability. Comparison pages show checked dates, limited-price periods, and public market sources.

Can I use OpenAI SDKs, Codex, and Claude Code?+

Yes—use an interface listed on the model page: Chat Completions or Responses for OpenAI SDKs, Responses for Codex, and Messages for Claude Code. The integration guide includes configuration examples.

How are API keys protected?+

Full secrets are shown once at creation; the server stores hashes. Create separate keys per application, set budgets and expiry dates, and revoke access when needed. Keep keys in server environment variables.