配置选项

from llama_cloud_services import LlamaExtract
from llama_cloud import ExtractConfig, ExtractMode, ChunkMode, ExtractTarget

# Initialize the client
extract = LlamaExtract(api_key="YOUR_API_KEY")

# Define your schema
schema = {
    "type": "object",
    "properties": {
        "company_name": {"type": "string", "description": "Name of the company"},
        "revenue": {"type": "number", "description": "Annual revenue in USD"}
    }
}

# Create extraction configuration
config = ExtractConfig(
    # Schema alignment
    extraction_target=ExtractTarget.PER_DOC,   # PER_DOC, PER_PAGE

    # Model settings
    extraction_mode=ExtractMode.MULTIMODAL,    # FAST, BALANCED, MULTIMODAL, PREMIUM
    parse_model=None,                          # Optional: override parse model
    extract_model=None,                        # Optional: override extract model

    # System prompt
    system_prompt="Focus on the most recent financial data",

    # Page range and context
    page_range="1-5,10-15",                    # Extract from specific pages
    num_pages_context=3,                       # Number of context pages for long docs

    # Metadata extensions (see Metadata Extensions page for details)
    cite_sources=True,                         # Enable citations
    use_reasoning=True,                        # Enable reasoning (not available in FAST mode)
    confidence_scores=True,                    # Enable confidence scores (MULTIMODAL/PREMIUM only)

    # Advanced options
    chunk_mode=ChunkMode.PAGE,                 # PAGE, SECTION
    high_resolution_mode=True,                 # Enable for better OCR
    invalidate_cache=False,                    # Set to True to bypass cache
)

# Create extraction agent with configuration
agent = extract.create_agent(
    name="Financial Data Extractor",
    data_schema=schema,
    config=config
)

# Extract from a document
result = agent.extract("your-document.pdf")
print(result.data)