Back to Models
Kimi K3 is Moonshot AI's successor to Kimi K2.6, verified live 2026-07-27. It targets the same agentic-coding niche as its predecessor with a 1M-token context window.
Best for
- • Teams already on the Kimi family who want the newest checkpoint for agentic coding
Reach for something else if
- • Vision/multimodal tasks — Kimi K3 is text-only in this catalog
Context Length
1M
Max Output
64K
Input Priceper 1M tokens
$3.24/ 1M tokens
Output Priceper 1M tokens
$16.20/ 1M tokens
Modalities
text→text
Pricing Breakdown
| Type | Rate |
|---|---|
| Input | $3.24 / 1M tokens |
| Output | $16.20 / 1M tokens |
Supported Parameters
temperaturemax_tokenstop_ptoolstool_choiceresponse_formatstop
Specifications
| Release date | Not publicly disclosed | unknown |
| License | Open-weight (per Moonshot's Kimi family convention) | to verify |
Benchmarks
| Benchmark | Distribution | Score | Source |
|---|---|---|---|
Not publicly disclosed | — | Not publicly disclosed | — |
API Usage Examples
Use the global api.therouter.ai endpoint shown below for new integrations; the legacy China accelerated endpoint is retired.
cURL
curl https://api.therouter.ai/v1/chat/completions -H "Content-Type: application/json" -H "Authorization: Bearer $THE_ROUTER_API_KEY" -d '{
"model": "moonshot/kimi-k3",
"messages": [
{"role": "user", "content": "Summarize the key points from this input."}
]
}'API guide
Chat completion
cURL
curl https://api.therouter.ai/v1/chat/completions \
-H "Authorization: Bearer $THEROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"moonshot/kimi-k3","messages":[{"role":"user","content":"Hello"}]}'More from moonshot
Guides
- GuideKimi K3 vs Qwen3.8-Max vs DeepSeek V4 Pro: Chinese Frontier Model API Comparison for Coding Agents (September 2026)
- GuideUsing Kimi K3 & DeepSeek V4.1 Flash via OpenAI-Compatible API: Integration Guide for TheRouter
- GuideKimi K3 vs Qwen3.8 Max vs Claude Opus 5: Frontier Reasoning Model API Comparison (August 2026)
- GuideKimi K3 API Complete Guide: 2.8T Reasoning Model, Pricing, and OpenAI-Compatible Integration (2026)
- GuideKimi K3 vs DeepSeek V4 vs Qwen3.7-Max: Chinese Reasoning Model API Comparison (July 2026)
- GuideKimi K3 vs K2: What Changed, What Broke, and When to Upgrade Your Routing
Fact ledger — every claim on this page traces here
| source | URL | retrieved | |
|---|---|---|---|
| Release date | — | — | unknown |
| License | — | — | to verify |
| Not publicly disclosed | www.moonshot.ai ↗ | 2026-07-27 | unknown |