Cut Your Claude API Bill by 90% with Prompt Caching and Batches

Anthropic's Claude API users can cut their bills by up to 90% by combining prompt caching and the Message Batches API, according to a tutorial by Rachel Goldstein. Prompt caching bills repeated context at 10% of the normal input rate, and the Batches API takes 50% off non-urgent requests, so a cached token inside a batch costs 5% of the list price. The tutorial uses the claude-opus-5 model, priced at $5 per million input tokens and $25 per million output tokens, and provides Python code to measure token costs and implement caching.

Cut Your Claude API Bill by 90% with Prompt Caching and Batches Measure real token costs, then stack prompt caching and the Batches API to slash them. Rachel Goldstein https://sourcefeed.dev/u/rachel goldstein What you'll build You'll take a working Claude API https://platform.claude.com/docs/en/ integration, measure exactly what its input tokens cost, then cut that bill twice: prompt caching https://platform.claude.com/docs/en/build-with-claude/prompt-caching bills repeated context at 10% of the normal input rate, and the Message Batches API https://platform.claude.com/docs/en/build-with-claude/batch-processing takes 50% off everything that doesn't need an immediate response. Stacked, a cached token inside a batch costs 5% of the list price. Prerequisites - Python 3.9+ tested on 3.13 0.120.2 current as of August 2026 anthropic Python SDK- An API key from the Claude Console https://platform.claude.com/ - Model: claude-opus-5 — $5 per million input tokens, $25 per million output tokens. Cache pricing verified against the official docs: writes cost 1.25× base input 2× for the 1-hour TTL , reads cost 0.1×. mkdir claude-costs && cd claude-costs python3 -m venv .venv && source .venv/bin/activate pip install anthropic==0.120.2 export ANTHROPIC API KEY="sk-ant-..." Step 1: Capture a cost baseline Most real integrations re-send a big static block on every request — product docs, a style guide, few-shot examples. That's the money leak. Save this as baseline.py : python import anthropic client = anthropic.Anthropic Stand-in for the static context your app re-sends on every request DOC LINE = "Acme Widget API reference, section {}: the flux endpoint accepts JSON " "payloads with a power field measured in gigawatts and returns a signed " "delivery manifest within 30 seconds.\n" SYSTEM = "Answer questions using only the reference below.\n\n" + "".join DOC LINE.format i for i in range 200 resp = client.messages.create model="claude-opus-5", max tokens=1024, system=SYSTEM, messages= {"role": "user", "content": "What unit is the power field measured in?"} , u = resp.usage print f"input={u.input tokens} cache write={u.cache creation input tokens} cache read={u.cache read input tokens}" Run python baseline.py . The system prompt is roughly 8,000 tokens, and input tokens confirms you're paying full price for all of them — about $0.04 per request, $40 per 1,000 requests, before output tokens. Step 2: Add a cache breakpoint Caching is a prefix match: the API renders tools → system → messages and caches everything up to a cache control marker. Any byte change earlier in the prefix invalidates it, so keep stable content first and volatile content the user's question after the breakpoint. Save as cached.py : python import anthropic client = anthropic.Anthropic DOC LINE = "Acme Widget API reference, section {}: the flux endpoint accepts JSON " "payloads with a power field measured in gigawatts and returns a signed " "delivery manifest within 30 seconds.\n" SYSTEM = "Answer questions using only the reference below.\n\n" + "".join DOC LINE.format i for i in range 200 def ask question: str - None: resp = client.messages.create model="claude-opus-5", max tokens=1024, system= { "type": "text", "text": SYSTEM, "cache control": {"type": "ephemeral"}, 5-minute TTL } , messages= {"role": "user", "content": question} , u = resp.usage print f"input={u.input tokens} cache write={u.cache creation input tokens} cache read={u.cache read input tokens}" ask "What unit is the power field measured in?" ask "How fast is the delivery manifest returned?" The first call writes the cache at 1.25× input price; every identical-prefix call within the 5-minute TTL each read refreshes it bills those tokens at 0.1× — the 90% cut. You get up to 4 breakpoints per request, and on claude-opus-5 a prefix must be at least 512 tokens to cache at all. Step 3: Batch the non-urgent work Evaluations, backfills, and nightly summarization don't need answers in seconds. The Batches API takes any list of Messages requests up to 100,000 requests or 256 MB per batch , processes them asynchronously — most batches finish within an hour — and bills all usage at 50% of standard prices. Caching works inside batches too; use the 1-hour TTL since processing usually outlives the 5-minute one. Save as batch.py : python import time import anthropic from anthropic.types.message create params import MessageCreateParamsNonStreaming from anthropic.types.messages.batch create params import Request client = anthropic.Anthropic DOC LINE = "Acme Widget API reference, section {}: the flux endpoint accepts JSON " "payloads with a power field measured in gigawatts and returns a signed " "delivery manifest within 30 seconds.\n" SYSTEM = "Answer questions using only the reference below.\n\n" + "".join DOC LINE.format i for i in range 200 shared system = { "type": "text", "text": SYSTEM, "cache control": {"type": "ephemeral", "ttl": "1h"}, } questions = "What unit is the power field measured in?", "What does the flux endpoint return?", "Summarize the reference in one sentence.", batch = client.messages.batches.create requests= Request custom id=f"q-{i}", params=MessageCreateParamsNonStreaming model="claude-opus-5", max tokens=1024, system=shared system, messages= {"role": "user", "content": q} , , for i, q in enumerate questions print f"Created {batch.id}, status: {batch.processing status}" while batch.processing status = "ended": time.sleep 30 batch = client.messages.batches.retrieve batch.id print f"status: {batch.processing status}" for result in client.messages.batches.results batch.id : if result.result.type == "succeeded": text = next b.text for b in result.result.message.content if b.type == "text" print f"{result.custom id}: {text}" else: print f"{result.custom id}: {result.result.type}" Results arrive in any order, so always key on custom id , never on position. Results stay downloadable for 29 days. Verify it works python cached.py should print token counts vary by a few : input=15 cache write=8021 cache read=0 input=14 cache write=0 cache read=8021 The second line is the win: 8,021 tokens billed at $0.50/MTok instead of $5. python batch.py should end with: Created msgbatch 013Zva2CMHLNnXjNJJKqJ2EF, status: in progress status: ended q-0: The power field is measured in gigawatts. q-1: The flux endpoint returns a signed delivery manifest. q-2: ... At 1,000 requests against this 8,000-token prompt, input costs drop from ~$40 uncached to ~$4 with caching, and ~$2 cached inside a batch — with output tokens also half price in the batch. Troubleshooting cache read input tokens is 0 on every request. Caching fails silently, never with an error. Two usual causes: the prefix is under the 512-token minimum for claude-opus-5 other models require 1,024–4,096 , or a byte differs between requests — a timestamp interpolated into the system prompt, an unsorted json.dumps , a tool list that changes order. Diff the rendered prompts of two consecutive requests. anthropic.AuthenticationError: Error code: 401 - {'type': 'error', 'error': {'type': 'authentication error', 'message': 'invalid x-api-key'}} — the key isn't reaching the client. Re-run export ANTHROPIC API KEY=... in the same shell as the script, and check for a stale key in .env or your shell profile overriding it. AttributeError: 'MessageBatchErroredResult' object has no attribute 'message' — you read result.result.message without checking the result type. Only succeeded results carry .message ; branch on result.result.type first errored results carry .error instead . Batch results come back expired. The batch hit the 24-hour processing ceiling before those requests ran — common under heavy demand or when a batch trips rate limits. Expired requests aren't billed; collect their custom id s and resubmit just those in a new batch. Next steps Point cached.py at your real system prompt and tool definitions — a breakpoint on the last system block caches tools and system together. Audit for silent cache invalidators timestamps, UUIDs, non-deterministic serialization before trusting your hit rate. Then look at the token counting endpoint https://platform.claude.com/docs/en/build-with-claude/token-counting to estimate costs pre-flight, and the rate limits page https://platform.claude.com/docs/en/api/rate-limits — batches have their own limits separate from the Messages API. Sources & further reading - Prompt caching https://platform.claude.com/docs/en/build-with-claude/prompt-caching — platform.claude.com - Batch processing https://platform.claude.com/docs/en/build-with-claude/batch-processing — platform.claude.com - Pricing https://platform.claude.com/docs/en/pricing — platform.claude.com - Anthropic Python SDK https://github.com/anthropics/anthropic-sdk-python — github.com Rachel Goldstein https://sourcefeed.dev/u/rachel goldstein · Dev Tools Editor Rachel has been embedded in the developer tooling ecosystem for nearly eight years, covering everything from IDE wars and package-manager drama to the quiet rise of AI-assisted coding. She has a soft spot for open-source maintainers and an unhealthy number of terminal emulators installed on a single laptop. Discussion 0 No comments yet Be the first to weigh in.