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

# Agent with Ollama Parser Model

This example demonstrates how to use an Ollama model as a parser for structured output, combining different models for generation and parsing.

## Code

```python parser_model_ollama.py theme={null}
import random
from typing import List

from agno.agent import Agent, RunOutput
from agno.models.ollama import Ollama
from agno.models.openai import OpenAIChat
from pydantic import BaseModel, Field
from rich.pretty import pprint


class NationalParkAdventure(BaseModel):
    park_name: str = Field(..., description="Name of the national park")
    best_season: str = Field(
        ...,
        description="Optimal time of year to visit this park (e.g., 'Late spring to early fall')",
    )
    signature_attractions: List[str] = Field(
        ...,
        description="Must-see landmarks, viewpoints, or natural features in the park",
    )
    recommended_trails: List[str] = Field(
        ...,
        description="Top hiking trails with difficulty levels (e.g., 'Angel's Landing - Strenuous')",
    )
    wildlife_encounters: List[str] = Field(
        ..., description="Animals visitors are likely to spot, with viewing tips"
    )
    photography_spots: List[str] = Field(
        ...,
        description="Best locations for capturing stunning photos, including sunrise/sunset spots",
    )
    camping_options: List[str] = Field(
        ..., description="Available camping areas, from primitive to RV-friendly sites"
    )
    safety_warnings: List[str] = Field(
        ..., description="Important safety considerations specific to this park"
    )
    hidden_gems: List[str] = Field(
        ..., description="Lesser-known spots or experiences that most visitors miss"
    )
    difficulty_rating: int = Field(
        ...,
        ge=1,
        le=5,
        description="Overall park difficulty for average visitor (1=easy, 5=very challenging)",
    )
    estimated_days: int = Field(
        ...,
        ge=1,
        le=14,
        description="Recommended number of days to properly explore the park",
    )
    special_permits_needed: List[str] = Field(
        default=[],
        description="Any special permits or reservations required for certain activities",
    )


agent = Agent(
    model=OpenAIChat(id="o3"),
    description="You help people plan amazing national park adventures and provide detailed park guides.",
    output_schema=NationalParkAdventure,
    parser_model=Ollama(id="Osmosis/Osmosis-Structure-0.6B"),
)

national_parks = [
    "Yellowstone National Park",
    "Yosemite National Park",
    "Grand Canyon National Park",
    "Zion National Park",
    "Grand Teton National Park",
    "Rocky Mountain National Park",
    "Acadia National Park",
    "Mount Rainier National Park",
    "Great Smoky Mountains National Park",
    "Rocky National Park",
]
# Get the response in a variable
run: RunOutput = agent.run(national_parks[random.randint(0, len(national_parks) - 1)])
pprint(run.content)
```

## Usage

<Steps>
  <Snippet file="create-venv-step.mdx" />

  <Step title="Install libraries">
    ```bash theme={null}
    pip install -U agno pydantic rich
    ```
  </Step>

  <Step title="Export your OpenAI API key">
    <CodeGroup>
      ```bash Mac/Linux theme={null}
        export OPENAI_API_KEY="your_openai_api_key_here"
      ```

      ```bash Windows theme={null}
        $Env:OPENAI_API_KEY="your_openai_api_key_here"
      ```
    </CodeGroup>
  </Step>

  <Step title="Create a Python file">
    Create a Python file and add the above code.

    ```bash theme={null}
    touch parser_model_ollama.py
    ```
  </Step>

  <Step title="Run Agent">
    <CodeGroup>
      ```bash Mac theme={null}
      python parser_model_ollama.py
      ```

      ```bash Windows   theme={null}
      python parser_model_ollama.py
      ```
    </CodeGroup>
  </Step>

  <Step title="Find All Cookbooks">
    Explore all the available cookbooks in the Agno repository. Click the link below to view the code on GitHub:

    <Link href="https://github.com/agno-agi/agno/tree/main/cookbook/agents/input_and_output" target="_blank">
      Agno Cookbooks on GitHub
    </Link>
  </Step>
</Steps>