Token Optimization Advanced

# BEFORE: 47 tokens
"I would like you to please analyze the following piece of text
and then provide me with a detailed summary of the main points
that are discussed in the text. The text is as follows:"

# AFTER: 9 tokens
"Summarize the main points of this text:"

# Savings: 80% fewer tokens, same result quality

# BEFORE: ~150 tokens
"You are a highly knowledgeable and experienced customer support
agent for our company. Your role is to help customers with their
questions and concerns. You should always be polite, professional,
and helpful. You should provide accurate information based on our
company policies. If you don't know the answer, you should let
the customer know and offer to escalate their issue."

# AFTER: ~40 tokens
"Customer support agent. Be polite and accurate. Follow company
policy. Escalate if unsure. Never fabricate information."

import hashlib, json

class ResponseCache:
    def __init__(self):
        self.cache = {}

    def get_key(self, messages):
        content = json.dumps(messages, sort_keys=True)
        return hashlib.sha256(content.encode()).hexdigest()

    def get(self, messages):
        key = self.get_key(messages)
        return self.cache.get(key)

    def set(self, messages, response):
        key = self.get_key(messages)
        self.cache[key] = response

cache = ResponseCache()

# Check cache before API call
cached = cache.get(messages)
if cached:
    response = cached  # Free! No tokens used.
else:
    response = client.messages.create(...)
    cache.set(messages, response)

# Anthropic Batch API example
from anthropic import Anthropic

client = Anthropic()

# Create a batch of requests
batch = client.messages.batches.create(
    requests=[
        {
            "custom_id": "request-1",
            "params": {
                "model": "claude-sonnet-4-20250514",
                "max_tokens": 1024,
                "messages": [{"role": "user", "content": "Summarize article 1..."}]
            }
        },
        {
            "custom_id": "request-2",
            "params": {
                "model": "claude-sonnet-4-20250514",
                "max_tokens": 1024,
                "messages": [{"role": "user", "content": "Summarize article 2..."}]
            }
        }
    ]
)
# Results arrive within hours at 50% discount

import logging

class TokenTracker:
    def __init__(self):
        self.total_input = 0
        self.total_output = 0
        self.request_count = 0

    def log_usage(self, response):
        self.total_input += response.usage.input_tokens
        self.total_output += response.usage.output_tokens
        self.request_count += 1
        logging.info(
            f"Request {self.request_count}: "
            f"in={response.usage.input_tokens} "
            f"out={response.usage.output_tokens} "
            f"cumulative_in={self.total_input} "
            f"cumulative_out={self.total_output}"
        )