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

# reason

> LLM inference — completions, summarization, extraction

The `reason` capability provides access to large language models for text generation, analysis, and reasoning tasks.

## Providers

| Provider  | Models                                     |
| --------- | ------------------------------------------ |
| OpenAI    | gpt-4o, gpt-4o-mini, gpt-4-turbo           |
| Anthropic | claude-sonnet-4-20250514, claude-3-5-haiku |

## Basic Usage

```typescript theme={null}
const result = await saturn.reason({
  prompt: 'Explain quantum computing in simple terms',
});

console.log(result.data.content);
// → "Quantum computing is a type of computation that..."
```

## Parameters

<ParamField body="prompt" type="string" required>
  The text prompt to send to the LLM.
</ParamField>

<ParamField body="provider" type="string">
  Specific provider to use. If omitted, Saturn chooses the best available.
</ParamField>

<ParamField body="model" type="string">
  Specific model to use. Requires `provider` to be set.
</ParamField>

<ParamField body="systemPrompt" type="string">
  Optional system prompt for context.
</ParamField>

<ParamField body="maxTokens" type="number">
  Maximum tokens in the response. Default varies by model.
</ParamField>

<ParamField body="temperature" type="number">
  Sampling temperature (0-2). Lower is more deterministic.
</ParamField>

## Response

```typescript theme={null}
interface ReasonResponse {
  data: {
    content: string;        // Generated text
    finishReason: string;   // 'stop', 'length', etc.
    usage: {
      promptTokens: number;
      completionTokens: number;
      totalTokens: number;
    };
  };
  metadata: {
    chargedUsdCents: number;
    provider: string;
    model: string;
    latencyMs: number;
    auditId: string;
  };
}
```

## Examples

### Choose a Provider

```typescript theme={null}
// Use Anthropic specifically
const result = await saturn.reason({
  prompt: 'Analyze this code for bugs...',
  provider: 'anthropic',
  model: 'claude-sonnet-4-20250514',
});
```

### With System Prompt

```typescript theme={null}
const result = await saturn.reason({
  systemPrompt: 'You are a helpful coding assistant.',
  prompt: 'How do I reverse a string in Python?',
});
```

### Control Output Length

```typescript theme={null}
const result = await saturn.reason({
  prompt: 'Write a haiku about programming',
  maxTokens: 50,
  temperature: 0.7,
});
```

## Pricing

Pricing is per-token and varies by model:

| Model                    | Input           | Output           |
| ------------------------ | --------------- | ---------------- |
| gpt-4o                   | \$2.50/M tokens | \$10.00/M tokens |
| gpt-4o-mini              | \$0.15/M tokens | \$0.60/M tokens  |
| claude-sonnet-4-20250514 | \$3.00/M tokens | \$15.00/M tokens |

<Info>
  Saturn quotes the cost before execution. The quote is based on estimated tokens. Actual cost may be slightly different based on actual usage.
</Info>

## Error Handling

```typescript theme={null}
import { SaturnPolicyDeniedError, SaturnProviderError } from '@saturn-pay/sdk';

try {
  await saturn.reason({ prompt: 'Very long prompt...' });
} catch (err) {
  if (err instanceof SaturnPolicyDeniedError) {
    // Budget exceeded or capability not allowed
    console.log('Policy error:', err.message);
  } else if (err instanceof SaturnProviderError) {
    // Upstream provider error
    console.log('Provider error:', err.message);
  }
}
```