Parsing Schema

• Product extraction: Get prices, names, ratings, and availability
• Content scraping: Extract articles, reviews, or listings
• Data collection: Structured data from tables, lists, or complex layouts
• API-like responses: Transform HTML into clean JSON objects

​

Two modes

1. Schema Mode
2. LLM Mode

​

Schema Mode

• Stable page structures
• Cost-sensitive applications (lower pricing)
• Precise control over data extraction
• High-volume, consistent extraction

1. Create a parser with CSS selectors mapping to your data model
2. Define data types and structure
3. Nimble extracts data exactly as specified

​

LLM Mode

• Pages with varying structures
• Quick setup without selector maintenance
• Complex extraction requirements
• Adaptive data extraction

1. Define a data model (optional) or describe what to extract
2. AI analyzes the page and identifies relevant data
3. Data is extracted and structured automatically

​

Quick comparison

​

Choosing the right mode

• You need low-cost, predictable extraction
• Page structure is stable and CSS selectors work reliably
• You want complete control over what gets extracted
• Processing high volumes with consistent structure

• You want zero-setup, hands-free extraction
• Pages have varying structures or layouts
• Maintaining CSS selectors is challenging
• Higher cost is acceptable for reduced maintenance

​

Usage examples

​

Schema Mode

from nimble import Nimble
from pydantic import BaseModel

class ProductInfo(BaseModel):
    name: str
    price: float
    rating: float
    in_stock: bool

nimble = Nimble(api_key="YOUR-API-KEY")

result = nimble.extract(
    url="https://www.example.com/product",
    schema=ProductInfo
)

print(result)

​

LLM Mode

from nimble import Nimble
from pydantic import BaseModel

class ProductInfo(BaseModel):
    name: str
    price: float
    rating: float
    in_stock: bool

nimble = Nimble(api_key="YOUR-API-KEY")

result = nimble.extract(
    url="https://www.example.com/product",
    schema=ProductInfo,
    schema_prompt="Extract product details including name, price, customer ratings, and stock availability"
)

print(result)

​

Combining with other features

from nimble import Nimble
from pydantic import BaseModel

class ProductInfo(BaseModel):
    name: str
    price: float
    rating: float

nimble = Nimble(api_key="YOUR-API-KEY")

result = nimble.extract({
    "url": "https://www.example.com",
    "render": True,
    "browser_actions": [
        {
            "infinite_scroll": {
                "duration": 5000
            }
        }
    ],
    "schema": ProductInfo
})

print(result)

​

Next steps