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

# Run Modes

> DeepRails has six different run modes that let you balance cost, latency, and accuracy across all APIs. Choose the intelligence level of the models behind your evaluations to best fit your needs.

export const DefinitionCard = ({children}) => {
  return <Card variant="secondary">
    <div style={{
    padding: '0.5rem',
    border: '5px solid var(--primary-light)',
    borderRadius: '0.5rem',
    fontSize: '1.3rem',
    lineHeight: '1.4',
    boxShadow: '0 0 10px 10px var(--primary-light)'
  }}>
        {children}
      </div>

</Card>;
};

<DefinitionCard>
  <strong>Run Modes</strong> control how DeepRails executes evaluations across <a href="/monitor">Monitor</a> and <a href="/defend">Defend</a>. Every run uses <strong>two different LLMs in parallel</strong> to reduce bias and improve accuracy. The selected Run Mode determines which models are selected — from compact cost-efficient models to advanced reasoning models — so you can optimize your workflow.
</DefinitionCard>

## Why Run Modes Matter

Not every task requires the same evaluation depth. A simple summarization prompt can be tested cost effectively with smaller models, while multi-step reasoning (math generation, chained steps, or multi-task prompts) benefits from reasoning-capable models. DeepRails' Run Modes let you tune this balance.

* **Always two models in parallel:** Every evaluation uses two distinct LLMs to generate scores, avoiding single-model bias.
* **Reasoning vs. non-reasoning models:** For complex prompts, modes that include reasoning models yield better accuracy and interpretability.
* **Available everywhere:** Run Modes function the same across the Monitor and Defend APIs on all plans.

## The Six Run Modes

<CardGroup cols={2}>
  <Card title={<span style={{ fontSize: '1.05rem', fontWeight: 700 }}>Super Fast</span>}>
    <strong>High-speed lightweight checks</strong>. Rapid evaluations with minimal overhead, ideal for large-scale screening.

    <div style={{ marginTop: '0.75rem', fontSize: '0.8rem', lineHeight: '1.6' }}>
      <div>Accuracy  ●●○○○</div>
      <div>Speed     ●●●●○</div>
    </div>
  </Card>

  <Card title={<span style={{ fontSize: '1.05rem', fontWeight: 700 }}>Fast (default)</span>}>
    <strong>Balanced speed and accuracy</strong>. The default mode — a strong balance of evaluation quality and throughput for most use cases.

    <div style={{ marginTop: '0.75rem', fontSize: '0.8rem', lineHeight: '1.6' }}>
      <div>Accuracy  ●●●●○</div>
      <div>Speed     ●●●○○</div>
    </div>
  </Card>

  <Card title={<span style={{ fontSize: '1.05rem', fontWeight: 700 }}>Precision</span>}>
    <strong>Deep multi-model analysis</strong>. Recommended for complex prompts that benefit from reasoning.

    <div style={{ marginTop: '0.75rem', fontSize: '0.8rem', lineHeight: '1.6' }}>
      <div>Accuracy  ●●●●○</div>
      <div>Speed     ●●○○○</div>
    </div>
  </Card>

  <Card title={<span style={{ fontSize: '1.05rem', fontWeight: 700 }}>Precision Codex</span>}>
    <strong>Code-optimized deep analysis</strong>. Recommended for code-based workflows that benefit from specialized code analysis.

    <div style={{ marginTop: '0.75rem', fontSize: '0.8rem', lineHeight: '1.6' }}>
      <div>Accuracy  ●●●●○</div>
      <div>Speed     ●●○○○</div>
    </div>
  </Card>

  <Card title={<span style={{ fontSize: '1.05rem', fontWeight: 700 }}>Precision Max</span>}>
    <strong>Exhaustive multi-pass verification</strong>. Two reasoning models in parallel — best for mission-critical use cases where accuracy outweighs cost or latency. **Note: Does not support streaming.**

    <div style={{ marginTop: '0.75rem', fontSize: '0.8rem', lineHeight: '1.6' }}>
      <div>Accuracy  ●●●●●</div>
      <div>Speed     ●○○○○</div>
      <div>Streaming  ✕</div>
    </div>
  </Card>

  <Card title={<span style={{ fontSize: '1.05rem', fontWeight: 700 }}>Precision Max Codex</span>}>
    <strong>Ultimate code-aware verification</strong>. Two reasoning models with Codex-optimized deep analysis for the highest accuracy on code-based workflows. **Note: Does not support streaming.**

    <div style={{ marginTop: '0.75rem', fontSize: '0.8rem', lineHeight: '1.6' }}>
      <div>Accuracy  ●●●●●</div>
      <div>Speed     ●○○○○</div>
      <div>Streaming  ✕</div>
    </div>
  </Card>
</CardGroup>

<Warning>
  **Streaming Constraints**: `precision_max` and `precision_max_codex` do NOT support streaming due to their multi-pass verification architecture. If your application requires real-time streaming responses, use `super_fast`, `fast`, `precision`, or `precision_codex` instead.
</Warning>

<Note>
  Choosing whether to use reasoning models is often part of the <strong>prompt engineering process</strong>. If your task involves multi-step logic, mathematics, or complex instructions, <strong>Precision</strong> or <strong>Precision Max</strong> are recommended. If your task requires complex code analysis, use <strong>Precision Codex</strong> or <strong>Precision Max Codex</strong>.
</Note>

## Choosing the Right Run Mode

### At a Glance

<div data-page="auto-tables">
  <table className="run-modes-table">
    <thead>
      <tr>
        <th>Run Mode</th>
        <th>ID</th>
        <th>Accuracy</th>
        <th>Speed</th>
        <th>Streaming</th>
        <th>Web / File Search</th>
      </tr>
    </thead>

    <tbody>
      <tr>
        <td><strong>Super Fast</strong></td>
        <td><code>super\_fast</code></td>
        <td>●●○○○</td>
        <td>●●●●○</td>
        <td>Yes</td>
        <td>No</td>
      </tr>

      <tr>
        <td><strong>Fast (default)</strong></td>
        <td><code>fast</code></td>
        <td>●●●●○</td>
        <td>●●●○○</td>
        <td>Yes</td>
        <td>Yes</td>
      </tr>

      <tr>
        <td><strong>Precision</strong></td>
        <td><code>precision</code></td>
        <td>●●●●○</td>
        <td>●●○○○</td>
        <td>Yes</td>
        <td>Yes</td>
      </tr>

      <tr>
        <td><strong>Precision Codex</strong></td>
        <td><code>precision\_codex</code></td>
        <td>●●●●○</td>
        <td>●●○○○</td>
        <td>Yes</td>
        <td>Yes</td>
      </tr>

      <tr>
        <td><strong>Precision Max</strong></td>
        <td><code>precision\_max</code></td>
        <td>●●●●●</td>
        <td>●○○○○</td>
        <td>No</td>
        <td>Yes</td>
      </tr>

      <tr>
        <td><strong>Precision Max Codex</strong></td>
        <td><code>precision\_max\_codex</code></td>
        <td>●●●●●</td>
        <td>●○○○○</td>
        <td>No</td>
        <td>Yes</td>
      </tr>
    </tbody>
  </table>
</div>

### Detailed Comparison

<div data-page="auto-tables">
  <table className="run-modes-table">
    <thead>
      <tr>
        <th>Name</th>
        <th>Description</th>
        <th>When to Use</th>
        <th>Example Use Case</th>
      </tr>
    </thead>

    <tbody>
      <tr>
        <td><strong>Super Fast</strong></td>
        <td>High-speed lightweight checks; rapid evaluations with minimal overhead.</td>
        <td>Large-scale screening, early exploration, low-stakes triage.</td>
        <td>Screening 10,000 code-gen outputs to flag potential safety risks.</td>
      </tr>

      <tr>
        <td><strong>Fast (default)</strong></td>
        <td>Balanced speed and accuracy; the default mode for most use cases.</td>
        <td>General-purpose workflows that need a good balance of quality and throughput.</td>
        <td>Monitoring daily regressions in a customer support chatbot.</td>
      </tr>

      <tr>
        <td><strong>Precision</strong></td>
        <td>Deep multi-model analysis; strong reasoning coverage with balanced cost/latency.</td>
        <td>Complex prompts with logic/calculations or multi-step reasoning.</td>
        <td>Monitoring daily regressions in a legal research bot.</td>
      </tr>

      <tr>
        <td><strong>Precision Codex</strong></td>
        <td>Code-optimized deep analysis; specialized for code-based workflows.</td>
        <td>Code reviews, software development tasks, and technical documentation requiring code analysis.</td>
        <td>Evaluating AI-generated code snippets for correctness and best practices.</td>
      </tr>

      <tr>
        <td><strong>Precision Max</strong></td>
        <td>Exhaustive multi-pass verification; two reasoning models in parallel (very high cost and latency).</td>
        <td>Mission-critical evaluations, final QA sweeps, regulated or safety-sensitive domains.</td>
        <td>Compliance evaluation on a healthcare agent before production.</td>
      </tr>

      <tr>
        <td><strong>Precision Max Codex</strong></td>
        <td>Ultimate code-aware verification (highest accuracy, lowest speed).</td>
        <td>Complex code creation or refactoring, debugging large systems, or similar software engineering tasks.</td>
        <td>Final security and correctness review of AI-generated code before merging to production.</td>
      </tr>
    </tbody>
  </table>
</div>

<Warning>
  **Super Fast** does not support **Web Search** or **File Search** capabilities. If your workflow or monitor has these enabled, requests using `super_fast` will be rejected. To resolve this, either switch to a run mode that supports these capabilities or edit the workflow/monitor to disable Web Search and File Search. This mode does support **Context Awareness** — you can still pass context in your `model_input`.
</Warning>