Kimi K3 is documented as suited to long-horizon coding, including work across large repositories and long-running engineering tasks. It is also documented for knowledge work, especially workflows that require sustained context across large documents and related materials. The model is documented for deep reasoning tasks and reasoning-heavy workflows. It is additionally documented for visual understanding and multimodal document or image analysis.
import requests
r = requests.post(
"https://api.aimlapi.com/v1/chat/completions",
headers={"Authorization": "Bearer " + AIMLAPI_KEY},
json={
"model": "moonshot/kimi-k3",
"messages": [
{
"role": "user",
"content": "Hello!"
}
]
},
)
print(r.json())
const r = await fetch("https://api.aimlapi.com/v1/chat/completions", {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.AIMLAPI_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
"model": "moonshot/kimi-k3",
"messages": [
{
"role": "user",
"content": "Hello!"
}
]
}),
});
console.log(await r.json());
curl -X POST https://api.aimlapi.com/v1/chat/completions \
-H "Authorization: Bearer $AIMLAPI_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"moonshot/kimi-k3","messages":[{"role":"user","content":"Hello!"}]}'
OpenAI-compatible — swap the base URL and it works with your existing SDK.
| Type | Price |
|---|---|
| Input | |
| Output | |
| Cached input |
| Benchmark | Score | What it measures | Source | Retrieved |
|---|---|---|---|---|
| Intelligence | 43.8 | Composite score across standardised reasoning, knowledge and problem-solving evaluations, measured independently by Artificial Analysis | Source | September 12, 2026 |
| Coding | 76.2 | Composite score across standardised coding evaluations, measured independently by Artificial Analysis | Source | September 12, 2026 |
| Model | Input | Output | Context | Best for |
|---|---|---|---|---|
Kimi K3 This page | Long-context, multimodal & agentic workflows | |||
| Coding + agents | ||||
| Coding + agents | ||||
| Reasoning + agents | ||||
| Reasoning + agents |
Kimi K3 has a 1,000,000 tokens context window and can return up to 128,000 tokens.
Kimi K3 takes text, image, file in as input and returns text out.
Use moonshot/kimi-k3 as the model id.
Kimi K3 became available on July 16, 2026.
Kimi K3 is priced at input $3.9 / 1M tokens, output $19.5 / 1M tokens, cached input $0.39 / 1M tokens.
Yes, Kimi K3 can stream responses as they are generated.
Yes, Kimi K3 accepts image input alongside text.
Kimi K3 was built by Moonshot AI.
Kimi K3 offers reasoning, function calling, vision, web search, structured outputs, and streaming, making it useful for tasks that combine text and image understanding with tool use and search integration.
Yes, Kimi K3 supports parallel tool calling.
