Prompt Caching

Enable caching for long prompts that are reused. Later requests can reduce cost by about 90% and latency by about 85%.

Python

import anthropic

client = anthropic.Anthropic(api_key="your-api-key", base_url="https://aisupermarket.work")

with open("long_document.txt") as f:
    document = f.read()

response = client.messages.create(
    model="claude-opus-4-7",
    max_tokens=1024,
    system=[
        {"type": "text", "text": "You are a document analysis assistant."},
        {
            "type": "text",
            "text": document,
            "cache_control": {"type": "ephemeral"}
        }
    ],
    messages=[{"role": "user", "content": "Please summarize the main points of this document"}]
)

print(response.content[0].text)
print(f"Cache creation tokens: {response.usage.cache_creation_input_tokens}")
print(f"Cache read tokens: {response.usage.cache_read_input_tokens}")

JavaScript

import Anthropic from "@anthropic-ai/sdk";
import fs from "fs";

const client = new Anthropic({ apiKey: "your-api-key", baseURL: "https://aisupermarket.work" });

const document = fs.readFileSync("long_document.txt", "utf-8");

const response = await client.messages.create({
  model: "claude-opus-4-7",
  max_tokens: 1024,
  system: [
    { type: "text", text: "You are a document analysis assistant." },
    { type: "text", text: document, cache_control: { type: "ephemeral" } },
  ],
  messages: [{ role: "user", content: "Please summarize the main points of this document" }],
});

console.log(response.content[0].text);
console.log(`Cache creation tokens: ${response.usage.cache_creation_input_tokens}`);
console.log(`Cache read tokens: ${response.usage.cache_read_input_tokens}`);

curl

curl https://aisupermarket.work/v1/messages \
  -H "x-api-key: your-api-key" \
  -H "anthropic-version: 2023-06-01" \
  -H "content-type: application/json" \
  -d '{
    "model": "claude-opus-4-7",
    "max_tokens": 1024,
    "system": [
      {"type": "text", "text": "You are a document analysis assistant."},
      {"type": "text", "text": "<your long document here>", "cache_control": {"type": "ephemeral"}}
    ],
    "messages": [{"role": "user", "content": "Please summarize the main points of this document"}]
  }'

Cache Tool Definitions in Multi-Turn Chat

Python

import anthropic

client = anthropic.Anthropic(api_key="your-api-key", base_url="https://aisupermarket.work")

tools = [
    {
        "name": "search",
        "description": "Search the knowledge base",
        "input_schema": {
            "type": "object",
            "properties": {"query": {"type": "string"}},
            "required": ["query"]
        },
        "cache_control": {"type": "ephemeral"}
    }
]

response = client.messages.create(
    model="claude-opus-4-7",
    max_tokens=1024,
    tools=tools,
    messages=[{"role": "user", "content": "Search for content about Python"}]
)
print(response.content[0].text)

JavaScript

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({ apiKey: "your-api-key", baseURL: "https://aisupermarket.work" });

const tools = [
  {
    name: "search",
    description: "Search the knowledge base",
    input_schema: {
      type: "object",
      properties: { query: { type: "string" } },
      required: ["query"],
    },
    cache_control: { type: "ephemeral" },
  },
];

const response = await client.messages.create({
  model: "claude-opus-4-7",
  max_tokens: 1024,
  tools,
  messages: [{ role: "user", content: "Search for content about Python" }],
});
console.log(response.content[0].text);

curl

curl https://aisupermarket.work/v1/messages \
  -H "x-api-key: your-api-key" \
  -H "anthropic-version: 2023-06-01" \
  -H "content-type: application/json" \
  -d '{
    "model": "claude-opus-4-7",
    "max_tokens": 1024,
    "tools": [{
      "name": "search",
      "description": "Search the knowledge base",
      "input_schema": {"type": "object", "properties": {"query": {"type": "string"}}, "required": ["query"]},
      "cache_control": {"type": "ephemeral"}
    }],
    "messages": [{"role": "user", "content": "Search for content about Python"}]
  }'

The minimum cache lifetime is 5 minutes and the maximum is 1 hour. At least 1024 tokens must be cached (2048 for Haiku).