Kimi K3 API
Model ID: kimi-k3 · Developer: Moonshot AI · Available on 7API: yes, as of 2026-11-20
Call Kimi K3 with any OpenAI- or Anthropic-compatible client. Use base URL https://7api.st/v1 (OpenAI) or https://7api.st (Anthropic) and send "model": "kimi-k3".
Specifications
| Model ID | kimi-k3 |
| Context window | 1,000,000 tokens |
| Max output | 65,536 tokens |
| Input | Text |
| Tool calling | Yes |
| Streaming | Yes |
| Reasoning effort | low, high, max (default max) |
| API formats | Chat Completions, Responses, Anthropic Messages |
Pricing
| USD per 1M tokens | |
|---|---|
| Input | $0.15 |
| Output | $0.15 |
| Cached input | $0.03 |
For example, an agent turn with 150,000 cached input tokens, 2,000 new input tokens and 800 output tokens costs about $0.0049 on Kimi K3. Billing is pay per token from a prepaid balance. See Pricing and billing.
Example
from openai import OpenAI
client = OpenAI(base_url="https://7api.st/v1", api_key="YOUR_7API_KEY")
r = client.chat.completions.create(
model="kimi-k3",
messages=[{"role": "user", "content": "Write a haiku about caching."}],
)
print(r.choices[0].message.content)In Claude Code: /model claude-kimi-k3. Setup guides: Claude Code, OpenCode, Codex, Cursor.
Related models
- Kimi K3.1 Pro: $0.30/M, $0.05/M cached, 1M context
