Name: Kimi K2.6 API: Coding and Reasoning LLM by Moonshot AI
Brand: magicapi
SKU: api-magicapi-ai-software-engineering-moonshotai-kimi-k2-6
Availability: InStock
Author: API Market

Shashank Agarwal

# 🤖 Coding and Reasoning LLM API (Moonshot AI Kimi K2.6)

## Overview
Kimi K2.6 is a large language model from Moonshot AI. Its developer describes it as a frontier open model built for long-horizon coding, agent swarms and autonomous software engineering, with 1 trillion parameters, a 262k context window, vision and tool use.

Through this API you send a text `prompt`, an optional `system_prompt` and optional images in `image_input`; the model returns generated text. `reasoning_effort` sets how much the model reasons before answering, and `max_tokens` (up to 32,768) caps the answer length. The input schema has no field for tool definitions, so tool calling is not available through this API.

## Features
- **Image input**: pass a list of image URLs in `image_input` to include them in the prompt.
- **Reasoning effort**: `reasoning_effort` accepts `none` (default, reasoning disabled), `low`, `medium` or `high`.
- **Long answers**: `max_tokens` from 1 to 32,768 (default 2048).
- **System prompt**: `system_prompt` (default "You are a helpful assistant.") guides the model's behavior.
- **Sampling controls**: `temperature` (0 to 2, default 0.6) and `top_p` (0 to 1, default 1).
- **Repetition controls**: `presence_penalty` and `frequency_penalty` (-2 to 2, default 0).

## Use Cases
1. **Code generation**: functions, scripts and tests from a description, as in the example below.
2. **Code review and explanation**: paste code into the prompt and ask for a review, a refactor or an explanation.
3. **Screenshots and diagrams**: send a UI screenshot or a diagram in `image_input` and ask questions about it.
4. **Harder problems**: raise `reasoning_effort` for multi-step planning, debugging and math.

## Visual Examples
### Example Input
Prompt: "Write a Python function `group_anagrams(words)` that groups strings by their anagram equivalence class. Add a concise docstring, type hints, and a short example in an `if __name__ == '__main__':` block. Output just the code, no prose."

Settings: `max_tokens` 800, `reasoning_effort` `none`. The system prompt was left at its default.

### Example Output
The chunks of this run joined into one answer:

````markdown
```python
from typing import List, Dict
from collections import defaultdict

def group_anagrams(words: List[str]) -> Dict[str, List[str]]:
    """Group words by their anagram equivalence class.
    
    Returns a dict mapping sorted characters to list of anagrams.
    """
    groups = defaultdict(list)
    for word in words:
        key = "".join(sorted(word))
        groups[key].append(word)
    return dict(groups)

if __name__ == '__main__':
    words = ["eat", "tea", "tan", "ate", "nat", "bat"]
    result = group_anagrams(words)
    print(result)
    # {'aet': ['eat', 'tea', 'ate'], 'ant': ['tan', 'nat'], 'abt': ['bat']}
```
````

## Input Parameters
| Field | Type | Required | Default | Allowed values | Description |
| --- | --- | --- | --- | --- | --- |
| `prompt` | string | No | `""` |  | Text prompt to send to Kimi K2.6 |
| `system_prompt` | string | No | `You are a helpful assistant.` |  | System prompt to guide the model's behavior |
| `image_input` | array of strings (URLs) | No | `[]` |  | Optional list of image URLs to include in the prompt |
| `max_tokens` | integer | No | `2048` | 1 to 32768 | Maximum number of tokens to generate |
| `temperature` | number | No | `0.6` | 0 to 2 | Sampling temperature. Higher values make output more random. |
| `top_p` | number | No | `1` | 0 to 1 | Nucleus sampling probability threshold |
| `presence_penalty` | number | No | `0` | -2 to 2 | Penalizes new tokens based on whether they appear in the text so far |
| `frequency_penalty` | number | No | `0` | -2 to 2 | Penalizes new tokens based on their existing frequency in the text so far |
| `reasoning_effort` | string | No | `none` | `none`, `low`, `medium`, `high` | How much effort the model spends reasoning before answering. Use `none` to disable. |

## Pricing
- **Pay As You Go (PAYG) pricing**: you only pay for what you use.
- **Price**: $0.001 per API unit.
- **1 API unit = 100 weighted tokens**; an output token counts 1, an input token counts 0.2375.
- That is **$2.375 per million input tokens** and **$10 per million output tokens**.
- Each request is rounded up to a whole API unit.
- **Example cost**: the request above used 76 input tokens and 168 output tokens. 76 x 0.2375 + 168 = 186.05 weighted tokens = 1.86 units, rounded up to 2 API units = **$0.002**.
- **Failed or cancelled predictions are not billed.**
- **GET requests to check prediction status** are free and don't count toward API unit usage.
- **No monthly subscription fee.**

## API Documentation

### Base URL
All API requests go to: `https://prod.api.market/api/v1/magicapi/ai-software-engineering-moonshotai-kimi-k2-6`

### Authentication
Send your api.market API key in the `x-api-market-key` header:
```
x-api-market-key: YOUR_API_MARKET_KEY
```

### Asynchronous Processing
**IMPORTANT**: This API is asynchronous. The flow is:
1. Send a POST request to `/predictions`. A successful request returns **201 Created** with a prediction `id` and a `status` of `starting` or `processing`.
2. Poll GET `/predictions/{id}` with that ID.
3. Keep polling while `status` is `starting` or `processing`.
4. Stop when `status` is `succeeded` (read `output`), `failed` or `canceled` (read `error`).

A polling interval of 1 second and an overall timeout of 600 seconds (10 minutes) work well.

### Endpoints

#### Create a Prediction
```bash
curl -X POST "https://prod.api.market/api/v1/magicapi/ai-software-engineering-moonshotai-kimi-k2-6/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  -d '{
    "version": "moonshotai/kimi-k2.6",
    "input": {
      "prompt": "Write a Python function `group_anagrams(words)` that groups strings by their anagram equivalence class. Add a concise docstring, type hints, and a short example in an `if __name__ == '\''__main__'\'':` block. Output just the code, no prose.",
      "max_tokens": 800,
      "reasoning_effort": "none"
    }
  }'
```

Response (201 Created):
```json
{
  "id": "spxbsq7eznrmr0d1c71vgs3xnw",
  "version": "moonshotai/kimi-k2.6",
  "status": "starting",
  "created_at": "2026-10-22T03:59:53.277Z"
}
```

#### Check Prediction Status
```bash
curl -X GET "https://prod.api.market/api/v1/magicapi/ai-software-engineering-moonshotai-kimi-k2-6/predictions/{id}" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

Response when the prediction has finished (200 OK):
```json
{
  "id": "spxbsq7eznrmr0d1c71vgs3xnw",
  "status": "succeeded",
  "output": [
    "```python\nfrom typing import List, Dict\nfrom collections import defaultdict\n\ndef group_anagrams(words: List[str]) -> Dict[str, List[str]]:\n    \"\"\"Group words by their anagram equivalence class.\n    \n    Returns a dict mapping sorted characters to list of anagrams.\n    \"\"\"\n    groups = defaultdict(list)\n    for word in words:\n        key = \"\".join(sorted(word))\n        groups[key].append(word)\n    return dict(groups)\n\nif __name__ == '__main__':\n    words = [\"eat\", \"tea\", \"tan\", \"ate\", \"nat\", \"bat\"]\n    result = group_anagrams(words)\n    print(result)\n    # {'aet': ['eat', 'tea', 'ate'], 'ant': ['tan', 'nat'], 'abt': ['bat']}\n```"
  ],
  "error": null,
  "created_at": "2026-10-22T03:59:53.277Z",
  "started_at": "2026-10-22T03:59:53.290Z",
  "completed_at": "2026-10-22T04:00:22.384Z"
}
```

`output` is an array of text chunks. Join them in order to get the full answer. The example above shows the chunks from this run already joined into one string.

### Status Codes
- **201 Created**: prediction created (POST `/predictions`). This is the normal success response.
- **200 OK**: prediction details returned (GET `/predictions/{id}`).
- **400 Bad Request**: invalid input, for example a `max_tokens` above 32768 or a `reasoning_effort` that is not in the list.
- **401 Unauthorized**: missing or invalid API key.
- **404 Not Found**: no prediction with that ID.
- **429 Too Many Requests**: rate limit exceeded.
- **500 Internal Server Error** or **503 Service Unavailable**: temporary server problem; retry later.

## Rate Limits
All plans allow 10 requests per second. Requests above that return `429 Too Many Requests`; back off before retrying and honour any `Retry-After` header. Polling a prediction with GET is free but still counts toward the per-second limit, so poll every 1 to 5 seconds rather than in a tight loop.

## Error Handling
- Check the HTTP status code of every request and read the response body for the error message.
- A prediction can also end with `status` set to `failed`; the `error` field then explains why. Failed predictions are not billed.
- Retry 429, 500 and 503 responses with exponential backoff. Fix the input before retrying a 400.

## Implementation Guide
### Example Code

#### JavaScript
```javascript
const apiKey = "YOUR_API_MARKET_KEY";
const baseUrl = "https://prod.api.market/api/v1/magicapi/ai-software-engineering-moonshotai-kimi-k2-6";

async function createAndWaitForPrediction(input) {
  const response = await fetch(`${baseUrl}/predictions`, {
    method: "POST",
    headers: {
      "Content-Type": "application/json",
      "x-api-market-key": apiKey
    },
    body: JSON.stringify({ version: "moonshotai/kimi-k2.6", input })
  });

  if (response.status !== 201) {
    throw new Error(`API error: ${response.status} ${await response.text()}`);
  }

  const prediction = await response.json();
  return pollForResults(prediction.id);
}

const delay = (ms) => new Promise((resolve) => setTimeout(resolve, ms));

async function pollForResults(predictionId) {
  const maxTimeout = 600000;
  const pollInterval = 1000;
  const startTime = Date.now();

  while (Date.now() - startTime < maxTimeout) {
    const response = await fetch(`${baseUrl}/predictions/${predictionId}`, {
      headers: { "x-api-market-key": apiKey }
    });

    if (!response.ok) {
      throw new Error(`API error: ${response.status} ${await response.text()}`);
    }

    const prediction = await response.json();

    if (prediction.status === "succeeded") {
      return prediction.output.join("");
    }

    if (prediction.status === "failed" || prediction.status === "canceled") {
      throw new Error(`Prediction ${prediction.status}: ${prediction.error || "Unknown error"}`);
    }

    await delay(pollInterval);
  }

  throw new Error("Prediction timed out after 10 minutes");
}

createAndWaitForPrediction({
    "prompt": "Write a Python function `group_anagrams(words)` that groups strings by their anagram equivalence class. Add a concise docstring, type hints, and a short example in an `if __name__ == '__main__':` block. Output just the code, no prose.",
    "max_tokens": 800,
    "reasoning_effort": "none"
  }).then((output) => console.log(output));
```

#### Python
```python
import time
import requests

API_KEY = "YOUR_API_MARKET_KEY"
BASE_URL = "https://prod.api.market/api/v1/magicapi/ai-software-engineering-moonshotai-kimi-k2-6"

def create_and_wait_for_prediction(input_data):
    headers = {"Content-Type": "application/json", "x-api-market-key": API_KEY}
    response = requests.post(
        f"{BASE_URL}/predictions",
        headers=headers,
        json={"version": "moonshotai/kimi-k2.6", "input": input_data},
    )

    if response.status_code != 201:
        raise Exception(f"API error: {response.status_code} {response.text}")

    return poll_for_results(response.json()["id"])

def poll_for_results(prediction_id):
    max_timeout = 600
    poll_interval = 1
    headers = {"x-api-market-key": API_KEY}
    start_time = time.time()

    while time.time() - start_time < max_timeout:
        response = requests.get(f"{BASE_URL}/predictions/{prediction_id}", headers=headers)

        if response.status_code != 200:
            raise Exception(f"API error: {response.status_code} {response.text}")

        prediction = response.json()

        if prediction["status"] == "succeeded":
            return "".join(prediction["output"])

        if prediction["status"] in ("failed", "canceled"):
            raise Exception(f"Prediction {prediction['status']}: {prediction.get('error') or 'Unknown error'}")

        time.sleep(poll_interval)

    raise Exception("Prediction timed out after 10 minutes")

print(create_and_wait_for_prediction({
    "prompt": "Write a Python function `group_anagrams(words)` that groups strings by their anagram equivalence class. Add a concise docstring, type hints, and a short example in an `if __name__ == '__main__':` block. Output just the code, no prose.",
    "max_tokens": 800,
    "reasoning_effort": "none"
}))
```

## FAQ
- **Can I send images?**
  - Yes. Put one or more image URLs in `image_input`, and ask about them in `prompt`.
- **How do I make the model reason before answering?**
  - Set `reasoning_effort` to `low`, `medium` or `high`. The default, `none`, disables reasoning.
- **Can I pass tool definitions for function calling?**
  - No. This API accepts `prompt`, `system_prompt`, `image_input` and sampling settings only.
- **How long can an answer be?**
  - Up to 32,768 tokens, set with `max_tokens` (default 2048).
- **How is a request billed?**
  - By weighted tokens: input tokens count 0.2375 and output tokens count 1, at $0.001 per 100 weighted tokens, rounded up to a whole unit per request.
- **Is the 201 response from POST an error?**
  - No. 201 Created means the prediction was accepted. Use the returned `id` to poll GET `/predictions/{id}` until `status` is `succeeded`, `failed` or `canceled`.
- **What if my prediction fails?**
  - Failed predictions are not billed. Read the `error` field, adjust the input and try again.

## Contact Us
If you have any questions, reach out to us through the chat on the bottom right of the page. You can also email our support team at hello@magicapi.com.


Generate code and text with Moonshot AI's Kimi K2.6, with image input and adjustable reasoning effort. Pay as you go, no monthly fee.