zerox()

async def zerox(
    cleanup: bool = True,
    concurrency: int = 10,
    file_path: Optional[str] = "",
    image_density: int = 300,
    image_height: tuple[Optional[int], int] = (None, 1056),
    maintain_format: bool = False,
    model: str = "gpt-4o-mini",
    output_dir: Optional[str] = None,
    temp_dir: Optional[str] = None,
    custom_system_prompt: Optional[str] = None,
    select_pages: Optional[Union[int, Iterable[int]]] = None,
    **kwargs
) -> ZeroxOutput

from pyzerox import zerox

import asyncio
from pyzerox import zerox
import os

# Set up your API key
os.environ["OPENAI_API_KEY"] = "your-api-key"

async def main():
    result = await zerox(
        file_path="https://example.com/document.pdf",
        model="gpt-4o-mini"
    )
    print(result)

# Run the async function
result = asyncio.run(main())

import asyncio
from pyzerox import zerox
import os

os.environ["OPENAI_API_KEY"] = "your-api-key"

async def main():
    result = await zerox(
        file_path="./path/to/document.pdf",
        model="gpt-4o-mini",
        output_dir="./output"  # Save markdown to file
    )
    return result

result = asyncio.run(main())

import asyncio
from pyzerox import zerox

async def main():
    # Process only pages 1, 3, and 5
    result = await zerox(
        file_path="document.pdf",
        model="gpt-4o-mini",
        select_pages=[1, 3, 5]
    )
    return result

result = asyncio.run(main())

import asyncio
from pyzerox import zerox

async def main():
    # Useful for documents with tables spanning multiple pages
    result = await zerox(
        file_path="document.pdf",
        model="gpt-4o-mini",
        maintain_format=True
    )
    return result

result = asyncio.run(main())

import os

os.environ["OPENAI_API_KEY"] = "your-api-key"

result = await zerox(
    file_path="document.pdf",
    model="gpt-4o-mini"  # or "gpt-4o"
)

import os

os.environ["AZURE_API_KEY"] = "your-azure-api-key"
os.environ["AZURE_API_BASE"] = "https://example-endpoint.openai.azure.com"
os.environ["AZURE_API_VERSION"] = "2023-05-15"

result = await zerox(
    file_path="document.pdf",
    model="azure/gpt-4o-mini"  # Format: azure/<deployment_name>
)

import os

os.environ["GEMINI_API_KEY"] = "your-gemini-api-key"

result = await zerox(
    file_path="document.pdf",
    model="gemini/gemini-1.5-flash"  # Format: gemini/<model_name>
)

import os

os.environ["ANTHROPIC_API_KEY"] = "your-anthropic-api-key"

result = await zerox(
    file_path="document.pdf",
    model="claude-3-opus-20240229"
)

import os

os.environ["AWS_ACCESS_KEY_ID"] = "your-access-key"
os.environ["AWS_SECRET_ACCESS_KEY"] = "your-secret-key"
os.environ["AWS_REGION_NAME"] = "us-east-1"

result = await zerox(
    file_path="document.pdf",
    model="bedrock/anthropic.claude-3-sonnet-20240229-v1:0"
)

import json

# Load service account credentials
with open('path/to/vertex_ai_service_account.json', 'r') as file:
    vertex_credentials = json.load(file)

vertex_credentials_json = json.dumps(vertex_credentials)

result = await zerox(
    file_path="document.pdf",
    model="vertex_ai/gemini-1.5-flash-001",
    vertex_credentials=vertex_credentials_json
)

from pyzerox import zerox, DEFAULT_SYSTEM_PROMPT

# Use custom prompt
custom_prompt = """
Convert this document to markdown.
Focus on extracting tables and charts.
Return only markdown with no explanations.
"""

result = await zerox(
    file_path="document.pdf",
    model="gpt-4o-mini",
    custom_system_prompt=custom_prompt
)

result = await zerox(
    file_path="document.pdf",
    model="gpt-4o-mini",
    
    # Processing options
    concurrency=5,              # Process 5 pages concurrently
    maintain_format=True,       # Maintain formatting across pages
    select_pages=[1, 2, 3],    # Process only specific pages
    
    # Image conversion options
    image_density=300,          # DPI for image conversion
    image_height=(None, 1056),  # Max height for images
    
    # File management
    output_dir="./output",      # Save markdown to file
    temp_dir="./temp",          # Custom temp directory
    cleanup=True,               # Clean up temp files after processing
    
    # Model customization
    custom_system_prompt="...",  # Override default prompt
    
    # Additional LiteLLM parameters
    temperature=0.1,
    max_tokens=4096
)