Getting Started

Get structured output from any LLM in three steps.

Install


pip install parsec-llm

Choose Your Provider

Pick an adapter for your preferred LLM provider:


# OpenAI
from parsec.models.adapters import OpenAIAdapter
adapter = OpenAIAdapter(api_key="your-key", model="gpt-4o-mini")
 
# Anthropic
from parsec.models.adapters import AnthropicAdapter
adapter = AnthropicAdapter(api_key="your-key", model="claude-3-5-sonnet-20241022")
 
# Gemini
from parsec.models.adapters import GeminiAdapter
adapter = GeminiAdapter(api_key="your-key", model="gemini-pro")

Enforce Structure

Define your schema and get validated output:


from parsec.validators import JSONValidator
from parsec.enforcement import EnforcementEngine
 
# Set up enforcement
validator = JSONValidator()
engine = EnforcementEngine(adapter, validator, max_retries=3)
 
# Define what you want
schema = {
    "type": "object",
    "properties": {
        "name": {"type": "string"},
        "age": {"type": "integer"},
        "email": {"type": "string"}
    },
    "required": ["name", "email"]
}
 
# Get structured output
result = await engine.enforce(
    "Extract: John Doe is 30, contact at john@example.com",
    schema
)
 
print(result.parsed_output)
# {"name": "John Doe", "age": 30, "email": "john@example.com"}

What Just Happened?

Enforcement Engine sent your prompt to the LLM
Validator checked the response against your schema
If invalid, the engine automatically retried with feedback
You got guaranteed valid output matching your schema

Next Steps

Model Adapters - Learn about OpenAI, Anthropic, and Gemini adapters
Validators - JSON schema and Pydantic validation options
Testing - Write tests for your structured outputs
Logging - Monitor performance and debug issues