ai-batch 0.0.1

Batch processing for Anthropic's Claude API with structured output

AI Batch

Python SDK for batch processing with structured output and citation mapping.

• 50% cost savings via Anthropic's batch API pricing
• Structured output with Pydantic models
• Field-level citations map results to source documents
• Type safety with full validation

Currently supports Anthropic Claude. OpenAI support coming soon.

Installation

pip install ai-batch

Quick Start

from ai_batch import batch_files
from pydantic import BaseModel

class Invoice(BaseModel):
    company_name: str
    total_amount: str
    date: str

# Process PDFs with structured output + citations
job = batch_files(
    files=["invoice1.pdf", "invoice2.pdf", "invoice3.pdf"],
    prompt="Extract the company name, total amount, and date.",
    model="claude-3-5-sonnet-20241022",
    response_model=Invoice,
    enable_citations=True
)

results = job.results()
citations = job.citations()

Output

Structured Results:

[
  Invoice(company_name="TechCorp Solutions Inc.", total_amount="$12,500.00", date="March 15, 2024"),
  Invoice(company_name="DataFlow Systems", total_amount="$8,750.00", date="March 18, 2024")
]

Field-Level Citations:

[
  {
    "company_name": [Citation(cited_text="TechCorp Solutions Inc.", start_page=1)],
    "total_amount": [Citation(cited_text="TOTAL: $12,500.00", start_page=2)],
    "date": [Citation(cited_text="Date: March 15, 2024", start_page=1)]
  },
  # ... one dict per result
]

Four Modes

Response Model	Citations	Returns
❌	❌	List of strings
✅	❌	List of Pydantic models
❌	✅	List of strings + flat citation list
✅	✅	List of Pydantic models + field citation dicts

# Mode 1: Text only
job = batch_files(files=["doc.pdf"], prompt="Summarize this")

# Mode 2: Structured only  
job = batch_files(files=["doc.pdf"], prompt="Extract data", response_model=MyModel)

# Mode 3: Text with citations
job = batch_files(files=["doc.pdf"], prompt="Analyze this", enable_citations=True)

# Mode 4: Structured with field citations
job = batch_files(files=["doc.pdf"], prompt="Extract data", 
                  response_model=MyModel, enable_citations=True)

Message Processing

For direct message processing:

from ai_batch import batch

messages = [
    [{"role": "user", "content": "Is this spam? You've won $1000!"}],
    [{"role": "user", "content": "Meeting at 3pm tomorrow"}],
]

job = batch(
    messages=messages,
    model="claude-3-haiku-20240307",
    response_model=SpamResult
)

results = job.results()

Setup

export ANTHROPIC_API_KEY="your-api-key"

Examples

• examples/citation_example.py - Basic citation usage
• examples/citation_with_pydantic.py - Structured output with citations
• examples/spam_detection.py - Email classification
• examples/pdf_extraction.py - PDF processing

Limitations

• Citations only work with flat Pydantic models (no nested models)
• PDFs require Sonnet models for best results
• Batch jobs are asynchronous - call job.results() when ready