> ## Documentation Index
> Fetch the complete documentation index at: https://docs.kymaapi.com/llms.txt
> Use this file to discover all available pages before exploring further.

# DeepSeek R1

> Kyma's strongest pure reasoning model for hard analysis, math, and multi-step logic.

## Overview

`deepseek-r1` is the right choice when answer quality depends on deep reasoning more than speed. It is slower than the default models, but it performs better on logic-heavy work, difficult analysis, and complex planning.

## Specs

| Field              | Value                                    |
| ------------------ | ---------------------------------------- |
| Model ID           | `deepseek-r1`                            |
| Best for           | Reasoning, analysis, math, hard planning |
| Context window     | 64K                                      |
| Max output tokens  | 32K                                      |
| Input modalities   | Text                                     |
| Output modalities  | Text                                     |
| Tool calling       | Yes                                      |
| Structured outputs | Yes                                      |
| Prompt caching     | No                                       |
| Speed              | Slow                                     |
| Cost band          | Balanced                                 |
| Release stage      | Stable                                   |

## Use this when

* You are solving difficult logic or analysis tasks.
* You want the strongest reasoning path in the catalog.
* You can tolerate slower responses in exchange for better thinking quality.

## Pick something else when

* You want a better default for day-to-day work: use [`qwen-3.6-plus`](/models/qwen-3.6-plus).
* You need tool-heavy autonomous coding: use [`kimi-k2.6`](/models/kimi-k2.6).
* You need low latency or large throughput: use `qwen-3-32b` or `gemini-2.5-flash`.

## Example

```python theme={null}
from openai import OpenAI

client = OpenAI(base_url="https://kymaapi.com/v1", api_key="ky-...")

response = client.chat.completions.create(
    model="deepseek-r1",
    messages=[
        {"role": "user", "content": "Find the failure modes in this distributed queue design."}
    ]
)
```

## Agent query example

```bash theme={null}
curl "https://kymaapi.com/v1/models?reasoning=true&recommended_for=reasoning&release_stage=stable"
```

## Related aliases

| Alias       | Resolves to   |
| ----------- | ------------- |
| `reasoning` | `deepseek-r1` |