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

# Pricing

> Credit-based pricing for Orion agents.

## Agent Pricing

Orion agents use a variety of specialized capabilities for image, video, and document processing. Each capability has a credit cost based on the processing tier selected. The current conversion rate is:
<p align="center"><strong>100 credits = \$1</strong></p>

## Pricing by Category

Agent capabilities are priced based on the computational resources required. Select between **Fast**, **Auto**, and **Pro** tiers depending on your use case. These tiers apply to both Orion-1 and Orion-2 model families.

* **Auto** (`vlmrun-orion-1:auto` / `vlmrun-orion-2:auto`) uses the `Fast` tier tools by default, and is recommended for prototyping and high-volume processing. The agent will automatically select the `Pro` tier tools for more complex tasks. This is the default tier for Orion agents.
* **Fast** (`vlmrun-orion-1:fast` / `vlmrun-orion-2:fast`) uses the `Fast` tier tools, and is recommended for speed and cost-efficiency, using lighter models that process requests quickly.
* **Pro** (`vlmrun-orion-1:pro` / `vlmrun-orion-2:pro`) uses the `Pro` tier tools, and is recommended for production use cases requiring maximum quality, using more powerful models that produce higher quality results but take longer and cost more.

### Image

Image capabilities are priced **credits per image**.

<table align="center">
  <thead>
    <tr>
      <th align="left" width="20%" style={{ whiteSpace: 'nowrap' }}>Capability</th>
      <th width="50%">Description</th>
      <th width="12%" align="center">Fast</th>
      <th width="12%" align="center">Pro</th>
    </tr>
  </thead>

  <tbody>
    <tr>
      <td>Caption & Tag</td>
      <td>Generate captions and tags for images</td>
      <td align="center">1</td>
      <td align="center">2</td>
    </tr>

    <tr>
      <td>Detect, Point, Segment</td>
      <td>Detect objects, point to objects, and segment regions</td>
      <td align="center">1</td>
      <td align="center">2</td>
    </tr>

    <tr>
      <td>UI Parsing</td>
      <td>Parse UI elements from screenshots and application images</td>
      <td align="center">1</td>
      <td align="center">2</td>
    </tr>

    <tr>
      <td>Generate & Edit</td>
      <td>Generate or edit images from text prompts</td>
      <td align="center">4</td>
      <td align="center">24</td>
    </tr>

    <tr>
      <td>Tools</td>
      <td>Image I/O, rotation, cropping, and other utilities</td>
      <td align="center">Free</td>
      <td align="center">Free</td>
    </tr>
  </tbody>
</table>

### Document

Document capabilities are priced **credits per page**.

<table align="center">
  <thead>
    <tr>
      <th align="left" width="20%" style={{ whiteSpace: 'nowrap' }}>Capability</th>
      <th width="50%">Description</th>
      <th width="12%" align="center">Fast</th>
      <th width="12%" align="center">Pro</th>
    </tr>
  </thead>

  <tbody>
    <tr>
      <td>OCR, Layout</td>
      <td>Extract text and detect layout elements from documents</td>
      <td align="center">0.10 / page</td>
      <td align="center">0.40 / page</td>
    </tr>

    <tr>
      <td>Parsing</td>
      <td>Parse and extract structured data from documents</td>
      <td align="center">1 / page</td>
      <td align="center">2 / page</td>
    </tr>

    <tr>
      <td>Tools</td>
      <td>Document I/O, page navigation, and other utilities</td>
      <td align="center">0</td>
      <td align="center">0</td>
    </tr>
  </tbody>
</table>

### Video

Video capabilities are priced **credits per hour of input video** or **credits per second of generated output video**. The standard output is approximately 6 seconds.

<table align="center">
  <thead>
    <tr>
      <th align="left" width="20%" style={{ whiteSpace: 'nowrap' }}>Capability</th>
      <th width="50%">Description</th>
      <th width="12%" align="center">Fast</th>
      <th width="12%" align="center">Pro</th>
    </tr>
  </thead>

  <tbody>
    <tr>
      <td>Caption, Summary, & Tag</td>
      <td>Generate captions, summaries, and tags for video content</td>
      <td align="center">20 / hour</td>
      <td align="center">80 / hour</td>
    </tr>

    <tr>
      <td>Generate & Edit</td>
      <td>Generate video from text prompts and optional input images</td>
      <td align="center">15 / sec</td>
      <td align="center">40 / sec</td>
    </tr>

    <tr>
      <td>Tools</td>
      <td>Video I/O, sampling, trimming, and other utilities</td>
      <td align="center">0</td>
      <td align="center">0</td>
    </tr>
  </tbody>
</table>

<Note>
  Standard video generation produces approximately 6 seconds of video, resulting in approximately 240 credits for Pro tier (40 cr/s x 6s) = \$2.40  (at \$1 per 100 credits).
</Note>

***

## Service Tiers

Orion agent executions and chat completions accept an optional `service_tier` that scales the computed credit cost by a fixed multiplier — independent of the Fast/Auto/Pro model tier. The `service_tier` controls **delivery** (routing, latency, queue priority); the Fast/Auto/Pro tier controls **model quality**.

<table align="center">
  <thead>
    <tr>
      <th align="left">Tier</th>
      <th align="center">Multiplier</th>
      <th>When to use</th>
    </tr>
  </thead>

  <tbody>
    <tr>
      <td><code>standard</code> <em>(default)</em></td>
      <td align="center">1.0×</td>
      <td>Baseline rates and latency. Used when <code>service\_tier</code> is omitted, <code>null</code>, <code>"auto"</code>, or <code>"default"</code>.</td>
    </tr>

    <tr>
      <td><code>flex</code></td>
      <td align="center">0.5× <strong>(50% off)</strong></td>
      <td>Batch / background agent runs that can tolerate higher, more variable latency.</td>
    </tr>

    <tr>
      <td><code>priority</code></td>
      <td align="center">1.8×</td>
      <td>User-facing agent interactions that require the lowest queue times.</td>
    </tr>
  </tbody>
</table>

The multiplier is applied to the **total** agent cost — base agent cost + all tool calls — so e.g. a 100-credit Pro agent run costs 50 credits at `flex` and 180 credits at `priority`.

<CodeGroup>
  ```python Python theme={"theme":{"light":"github-light","dark":"dark-plus"}}
  # Flex tier — 50% discount on the total agent cost, higher latency
  response = client.agent.execute(
      name="my-orion-agent",
      inputs={...},
      service_tier="flex",
  )
  ```

  ```javascript Node.js theme={"theme":{"light":"github-light","dark":"dark-plus"}}
  // Priority tier — 1.8x premium, lowest latency
  await client.agent.execute({
    name: "my-orion-agent",
    inputs: {...},
    serviceTier: "priority",
  });
  ```

  ```bash cURL theme={"theme":{"light":"github-light","dark":"dark-plus"}}
  # Flex tier — 50% discount on the total agent cost, higher latency
  curl -X POST https://api.vlm.run/v1/agent/execute \
    -H "Authorization: Bearer $VLMRUN_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{
      "name": "my-orion-agent",
      "inputs": {...},
      "service_tier": "flex"
    }'
  ```
</CodeGroup>

Omitting the field (or passing `"auto"`, `"default"`, or `null`) falls back to the server default, which is currently `standard`. See the [main pricing page](/pricing#service-tiers) for the full tier table.

***

## Total Cost Calculation

## Examples

These examples show how credits are calculated for common multi-step agent workflows using `vlmrun-orion-1:auto` (Auto/Fast) or `vlmrun-orion-1:pro` (Pro).

<AccordionGroup>
  <Accordion title="1. Create 3 highlights from a 10-minute video and trim">
    | Step                    | Capability             | Auto / Fast                    | Pro                             |
    | ----------------------- | ---------------------- | ------------------------------ | ------------------------------- |
    | Caption & analyze video | Caption & Tag (10 min) | 3.33 credits                   | 13.33 credits                   |
    | Trim 3 highlight clips  | Tools                  | 0 credits                      | 0 credits                       |
    | **Total**               |                        | **3.33 credits** (**\$0.033**) | **13.33 credits** (**\$0.133**) |
  </Accordion>

  <Accordion title="2. Create a 5-part storyboard with generated video">
    | Step                             | Capability                  | Auto / Fast                  | Pro                            |
    | -------------------------------- | --------------------------- | ---------------------------- | ------------------------------ |
    | Generate 5 storyboard images     | Generate & Edit (5x)        | 20 credits                   | 120 credits                    |
    | Generate 5 video clips (6s each) | Generate & Edit (30s total) | 450 credits                  | 1200 credits (**\$12.00**)     |
    | **Total**                        |                             | **470 credits** (**\$4.70**) | **1320 credits** (**\$13.20**) |
  </Accordion>

  <Accordion title="3. Detect and segment objects in an image, then visualize">
    | Step                        | Capability   | Auto / Fast                | Pro                        |
    | --------------------------- | ------------ | -------------------------- | -------------------------- |
    | Detect objects in image     | Detection    | 1 credits                  | 2 credits                  |
    | Segment detected objects    | Segmentation | 1 credits                  | 2 credits                  |
    | Apply visualization overlay | Tools        | 0 credits                  | 0 credits                  |
    | **Total**                   |              | **2 credits** (**\$0.02**) | **4 credits** (**\$0.04**) |
  </Accordion>

  <Accordion title="4. Process a 50-page document with OCR and parsing">
    | Step                   | Capability             | Auto / Fast                 | Pro                         |
    | ---------------------- | ---------------------- | --------------------------- | --------------------------- |
    | OCR & layout detection | OCR, Layout (50 pages) | 5 credits                   | 20 credits                  |
    | Parse structured data  | Parsing (50 pages)     | 5 credits                   | 20 credits                  |
    | **Total**              |                        | **10 credits** (**\$0.10**) | **40 credits** (**\$0.40**) |
  </Accordion>
</AccordionGroup>

***

## FAQ

<AccordionGroup>
  <Accordion title="What's the difference between Auto, Fast and Pro tiers?">
    **Auto** (`vlmrun-orion-1:auto`) and **Fast** (`vlmrun-orion-1:fast`) are optimized for speed and cost-efficiency, using lighter models that process requests quickly. **Pro** (`vlmrun-orion-1:pro`) uses more powerful models that produce higher quality results but take longer and cost more. Choose Auto/Fast for prototyping and high-volume processing, and Pro for production use cases requiring maximum quality.
  </Accordion>

  <Accordion title="How is the total cost of the agent execution priced?">
    The total cost of the agent execution is the sum of the credits used for the tools, and the cost of the orchestration. For non-integer credits, we round up to the nearest credit.
  </Accordion>

  <Accordion title="How is the cost of the chat portion priced (excluding tools)?">
    The cost of the chat completion is priced per token. This is merely a simple cost overhead for orchestration, which we do not charge for if the cost of the single chat completion is less than 1 credit.
  </Accordion>

  <Accordion title="Are there any free tools?">
    Yes, utility tools like image I/O, basic editing operations, video sampling/trimming, and document navigation are included at no additional credit cost when used within an agent session. However, the base agent session cost still applies.
  </Accordion>
</AccordionGroup>

## Get Started

Ready to build with Orion agents? Check out our [Agent Introduction](/agents/introduction) to get started.

<CardGroup cols={2}>
  <Card title="Agent Documentation" icon="robot" href="/agents/introduction">
    Learn about Orion agent capabilities and how to use them.
  </Card>

  <Card title="Main Pricing" icon="dollar-sign" href="/pricing">
    View pricing for all VLM Run services including API and agent sessions.
  </Card>
</CardGroup>