Alibaba Cloud Model Studio:Explicit Cache Best Practices

Key points

• Add "cache_control": {"type": "ephemeral"} to the message content you want to cache. All content from the beginning of the messages array up to that marker will be cached as a block.

• Cached content must be at least 1024 tokens.

• A single request supports up to 4 cache markers.

• Cache TTL is 5 minutes, automatically renewed on each hit.

• Tool definitions are part of the system prompt for caching purposes. If the tools change, the cache will not hit.

Quick start

The following example demonstrates the basic workflow: the first request creates a cache, and the second request hits it.

from openai import OpenAI
import os

client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    base_url="https://dashscope-intl.aliyuncs.com/compatible-mode/v1",
)

# Long text to cache (must exceed 1024 tokens)
long_text_content = "<Your Long Text Here>" * 400

def get_completion(user_input):
    messages = [
        {
            "role": "system",
            "content": [
                {
                    "type": "text",
                    "text": long_text_content,
                    # Cache marker: content from the start of messages to this point will be cached
                    "cache_control": {"type": "ephemeral"},
                }
            ],
        },
        {"role": "user", "content": user_input},
    ]
    completion = client.chat.completions.create(
        model="qwen3.7-max",
        messages=messages,
        extra_body={"enable_thinking": False},
    )
    return completion

# First request: creates cache
first = get_completion("Summarize the key points of this document")
print(f"Cache created: {first.usage.prompt_tokens_details.cache_creation_input_tokens}")
print(f"Cache hit: {first.usage.prompt_tokens_details.cached_tokens}")

# Second request: same system content, different question — hits cache
second = get_completion("What precautions are mentioned in the document?")
print(f"Cache created: {second.usage.prompt_tokens_details.cache_creation_input_tokens}")
print(f"Cache hit: {second.usage.prompt_tokens_details.cached_tokens}")

import anthropic
import os

client = anthropic.Anthropic(
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    base_url="https://dashscope-intl.aliyuncs.com/apps/anthropic",
)

# Long text to cache (must exceed 1024 tokens)
long_text_content = "<Your Long Text Here>" * 400

def get_completion(user_input):
    response = client.messages.create(
        model="qwen3.7-max",
        max_tokens=1024,
        system=[
            {
                "type": "text",
                "text": long_text_content,
                # Cache marker
                "cache_control": {"type": "ephemeral"},
            }
        ],
        messages=[
            {"role": "user", "content": user_input},
        ],
    )
    return response

# First request: creates cache
first = get_completion("Summarize the key points of this document")
print(f"Cache created: {first.usage.cache_creation_input_tokens}")
print(f"Cache hit: {first.usage.cache_read_input_tokens}")

# Second request: hits cache
second = get_completion("What precautions are mentioned in the document?")
print(f"Cache created: {second.usage.cache_creation_input_tokens}")
print(f"Cache hit: {second.usage.cache_read_input_tokens}")

Expected output:

Cache created: 2005
Cache hit: 0
Cache created: 0
Cache hit: 2005

The first request creates a cache block. The second request hits the cache because the system prompt content is identical. Cached tokens are billed at only 10% of the standard input price.

Verify cache status

Check the usage field in the response to confirm cache behavior:

• cache_creation_input_tokens: Number of tokens for which a new cache was created. A value greater than 0 means a new cache block was created.

• cached_tokens (OpenAI-compatible) or cache_read_input_tokens (Anthropic-compatible): Number of tokens that hit cache. A value greater than 0 means the cache was successfully hit.

Multi-turn conversations

Characteristics:

• Users interact with the model over multiple turns, each request carrying the full conversation history

• Typical use cases: customer service, knowledge Q&A, code assistants

Best practice: Add a cache_control marker to the last message in each request. Each turn hits the cache created by the previous turn (the conversation history), while creating a new cache that includes the current turn for the next round.

Example:

from openai import OpenAI
import os

client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    base_url="https://dashscope-intl.aliyuncs.com/compatible-mode/v1",
)

# System prompt: product manual (must exceed 1024 tokens)
product_manual = """You are the support assistant for "BaiLian SmartHome" smart home controller. Here is the complete product manual:

## Product Overview
BaiLian SmartHome is a whole-home smart controller supporting voice control, scene automation, and energy management...

## Installation Guide
1. Install at a central location with good WiFi coverage...
2. Connect the power adapter (5V/2A)...

## FAQ
Q: Cannot connect to WiFi? A: Make sure your router supports 2.4GHz...
""" * 80  # Repeat to exceed 1024 tokens

messages = [{"role": "system", "content": product_manual}]

def chat(user_input):
    # Key: add cache_control to the last user message
    messages.append({
        "role": "user",
        "content": [
            {
                "type": "text",
                "text": user_input,
                "cache_control": {"type": "ephemeral"},
            }
        ],
    })
    completion = client.chat.completions.create(
        model="qwen3.7-max",
        messages=messages,
        extra_body={"enable_thinking": False},
    )
    assistant_msg = completion.choices[0].message.content
    messages.append({"role": "assistant", "content": assistant_msg})

    usage = completion.usage
    created = usage.prompt_tokens_details.cache_creation_input_tokens
    cached = usage.prompt_tokens_details.cached_tokens
    print(f"  [Cache] Created: {created} tokens, Hit: {cached} tokens")
    return assistant_msg

# Simulate multi-turn conversation
print("User: What voice assistants does BaiLian SmartHome support?")
print(f"Agent: {chat('What voice assistants does BaiLian SmartHome support?')[:80]}...\n")

print("User: What if I cannot connect to WiFi?")
print(f"Agent: {chat('What if I cannot connect to WiFi?')[:80]}...\n")

print("User: How many devices can it control simultaneously?")
print(f"Agent: {chat('How many devices can it control simultaneously?')[:80]}...")

Expected output:

User: What voice assistants does BaiLian SmartHome support?
  [Cache] Created: 8739 tokens, Hit: 0 tokens
Agent: BaiLian SmartHome supports Tmall Genie, XiaoAi, Siri, and other voice assistants...

User: What if I cannot connect to WiFi?
  [Cache] Created: 151 tokens, Hit: 8739 tokens
Agent: For WiFi connectivity issues, try the following: 1. Confirm your router supports 2.4GHz...

User: How many devices can it control simultaneously?
  [Cache] Created: 101 tokens, Hit: 8890 tokens
Agent: BaiLian SmartHome can control up to 256 smart devices simultaneously...

Starting from the second turn, each request hits the cache from the previous turn (the conversation history), while creating a new cache that includes the current turn. The more turns in the conversation, the greater the savings.

Production Agent (multiple cache markers)

Characteristics:

• Long multi-turn conversations comprising: system prompt + skills/tools definitions + project context + user messages / tool calls

• Different sections change at different frequencies

• Typical use cases: AI coding assistants (Claude Code, OpenClaw), RAG-based Q&A systems

Best practice: Use multiple cache markers (up to 4) to pin content at different stability levels. Each marker must be on a separate message (different role) to serve as an independent breakpoint:

• System prompt — one marker (rarely changes)

• Skills/tools definitions — one marker (may change in combination)

• Project context — one marker (may switch or compress)

• User messages / tool calls — one marker (grows each turn)

Example: This example simulates a typical Agent architecture with 3 cache markers pinning the system persona and tools (marker 1), knowledge base (marker 2), and conversation history (marker 3). Note that the knowledge base is placed in a user message to ensure it has its own independent cache breakpoint — multiple system messages are merged internally and cannot serve as separate breakpoints:

from openai import OpenAI
import os

client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    base_url="https://dashscope-intl.aliyuncs.com/compatible-mode/v1",
)

# Layer 1: System persona (rarely changes)
system_persona = """You are the senior AI support agent for "Model Studio Electronics". Your guidelines:
1. Answer questions based on the knowledge base
2. For information not in the knowledge base, say "Let me transfer you to a human agent"
3. Maintain a professional and friendly tone
4. If the user is unhappy, apologize first then resolve the issue

Below is your complete service specification and script guide:
""" + "Detailed service specification..." * 200  # Ensure > 1024 tokens

# Layer 2: Tools/skills definitions (changes occasionally, e.g. when new features launch)
tools_description = """### Available Tools
- search_product(query): Search product information
- check_inventory(sku, color): Check stock status
- create_ticket(type, description): Create a support ticket
- transfer_to_human(reason): Transfer to a human agent

### Tool Usage Rules
1. When user asks about product details, use search_product first
2. When user asks about stock/shipping, use check_inventory
3. When user requests return/exchange, use create_ticket
4. When a tool returns an error, apologize and transfer_to_human
""" + "Detailed tool usage examples..." * 150  # Ensure > 1024 tokens

# Layer 3: Project knowledge base (semi-stable, changes when user switches products)
knowledge_base_product_a = """### Current product: Model Studio Pro Max Wireless Earbuds
- SKU: BL-PM-2024
- Price: CNY 599
- Colors: Night Black / Nebula White / Ice Blue
- Battery: 8 hours (ANC on), 12 hours (ANC off)
- Water resistance: IPX5
- Warranty: 1 year, 7-day no-questions-asked return
- Stock: Night Black (in stock) / Nebula White (low) / Ice Blue (out of stock)
""" * 50  # Ensure > 1024 tokens

def ask_agent(user_question, history=None):
    if history is None:
        history = []
    messages = [
        {
            "role": "system",
            "content": [
                {
                    "type": "text",
                    "text": system_persona + "\n\n" + tools_description,
                    "cache_control": {"type": "ephemeral"},  # Marker 1: system persona + tools
                }
            ],
        },
        {
            "role": "user",
            "content": [
                {
                    "type": "text",
                    "text": f"Here is the knowledge base for the current product:\n{knowledge_base_product_a}",
                    "cache_control": {"type": "ephemeral"},  # Marker 2: knowledge base
                }
            ],
        },
        {"role": "assistant", "content": "Got it. I have the product details ready. How can I help you?"},
    ]
    messages.extend(history)
    # Add current question with marker 3
    messages.append({
        "role": "user",
        "content": [
            {
                "type": "text",
                "text": user_question,
                "cache_control": {"type": "ephemeral"},  # Marker 3: conversation history
            }
        ],
    })

    completion = client.chat.completions.create(
        model="qwen3.7-max",
        messages=messages,
        extra_body={"enable_thinking": False},
    )
    usage = completion.usage
    print(f"  Created: {usage.prompt_tokens_details.cache_creation_input_tokens}, "
          f"Hit: {usage.prompt_tokens_details.cached_tokens}")
    return completion.choices[0].message.content

# First request
print("Q1: Is the Ice Blue color available?")
a1 = ask_agent("Is the Ice Blue color available?")
print(f"A1: {a1}\n")

# Second request: same product (persona + tools + knowledge base all hit)
history = [
    {"role": "user", "content": "Is the Ice Blue color available?"},
    {"role": "assistant", "content": a1},
]
print("Q2: When will it be back in stock?")
a2 = ask_agent("When will it be back in stock?", history)
print(f"A2: {a2}")

Expected output:

Q1: Is the Ice Blue color available?
  Created: 7659, Hit: 0
A1: I'm sorry, but the Ice Blue color... is currently out of stock...

Q2: When will it be back in stock?
  Created: 73, Hit: 7659
A2: I don't have access to specific restock dates... Let me transfer you to a human agent...

In Q2, the prefix up to marker 2 (persona + tools + knowledge base = 7,659 tokens) is unchanged, resulting in a full cache hit. Only the new content after marker 2 (conversation history + new question) requires processing.

How multi-marker caching works:

• User continues asking about the same product: Persona, tools, and knowledge base are all unchanged, hitting the cache at marker 2 (longest prefix match) for maximum savings.

• More conversation turns: Earlier content (persona + tools + knowledge base + history) hits the previous turn's cache; only the new content requires a fresh cache.

Batch processing (task completion)

Characteristics:

• Single-turn requests, no context memory needed

• Fixed long system prompt (task instructions) + variable user input (data to process)

• Typical use cases: text classification, intent recognition, data extraction, content moderation

Best practice: Add the cache_control marker only on the system prompt. All subsequent requests hit the cache as long as the system prompt remains unchanged.

Example:

from openai import OpenAI
import os

client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    base_url="https://dashscope-intl.aliyuncs.com/compatible-mode/v1",
)

# Long system prompt: detailed classification rules (must exceed 1024 tokens)
classification_prompt = """You are a product review classifier. Classify each review into one of these categories:
- Positive
- Negative
- Neutral
- Question
- Complaint

Output only the category name, nothing else.

Detailed classification rules and examples:
""" + """Rules:
1. Positive: Contains positive sentiment words (e.g., "great", "excellent", "recommend"), or expresses satisfaction.
2. Negative: Contains negative sentiment words (e.g., "terrible", "disappointed", "return"), or expresses dissatisfaction.
3. Neutral: No clear sentiment, merely states facts.
4. Question: Phrased as a question asking for product information.
5. Complaint: Expresses suggestions for improvement or lodges a complaint.
""" * 100

# Reviews to classify (simulating batch processing)
reviews = [
    "This product is amazing, great quality, highly recommended!",
    "Shipping took a week and the packaging was damaged",
    "Does this come in red? Does it run large or small?",
    "You should add more size options, medium is too big for me",
    "It's okay I guess, nothing special, does what it says",
]

print("=== Batch Classification (Explicit Cache) ===")
for i, review in enumerate(reviews):
    completion = client.chat.completions.create(
        model="qwen3.7-max",
        messages=[
            {
                "role": "system",
                "content": [
                    {
                        "type": "text",
                        "text": classification_prompt,
                        "cache_control": {"type": "ephemeral"},  # Cache classification rules
                    }
                ],
            },
            {"role": "user", "content": review},
        ],
    )
    result = completion.choices[0].message.content
    cached = completion.usage.prompt_tokens_details.cached_tokens
    created = completion.usage.prompt_tokens_details.cache_creation_input_tokens
    print(f"Review {i+1}: \"{review[:40]}...\" -> {result}")
    print(f"  Created: {created}, Hit: {cached}")

Expected output:

Review 1: "This product is amazing, great quality, ..." -> Positive
  Created: 10353, Hit: 0
Review 2: "Shipping took a week and the packaging w..." -> Negative
  Created: 0, Hit: 10353
Review 3: "Does this come in red? Does it run large..." -> Question
  Created: 0, Hit: 10353
Review 4: "You should add more size options, medium..." -> Complaint
  Created: 0, Hit: 10353
Review 5: "It's okay I guess, nothing special, does..." -> Neutral
  Created: 0, Hit: 10353

After the first request creates the cache, all subsequent requests hit it. When processing 1,000 items, 999 requests see a 90% reduction in input token cost.

Function Calling with cached tool definitions

Characteristics:

• Using Function Calling with a long list of tool definitions

• Tool definitions remain unchanged across requests

Best practice: The tools parameter content is part of the system prompt for caching. Ensure tool definitions are exactly identical across requests (same order, same field order, same structure), and add a cache_control marker to the message content.

from openai import OpenAI
import os

client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    base_url="https://dashscope-intl.aliyuncs.com/compatible-mode/v1",
)

# Long text to meet the 1024-token minimum
long_text_content = "<Your Code Here>" * 400

# Tool definitions: must be exactly identical across requests
tools = [
    {
        "type": "function",
        "function": {
            "name": "get_weather",
            "description": "Get the current weather for a given city",
            "parameters": {
                "type": "object",
                "properties": {
                    "city": {"type": "string", "description": "City name"},
                    "unit": {"type": "string", "enum": ["celsius", "fahrenheit"]}
                },
                "required": ["city"]
            }
        }
    },
    {
        "type": "function",
        "function": {
            "name": "search_flights",
            "description": "Search flights between two cities",
            "parameters": {
                "type": "object",
                "properties": {
                    "origin": {"type": "string", "description": "Departure city"},
                    "destination": {"type": "string", "description": "Destination city"},
                    "date": {"type": "string", "description": "Departure date in YYYY-MM-DD format"}
                },
                "required": ["origin", "destination", "date"]
            }
        }
    }
]

def ask(user_input):
    messages = [
        {
            "role": "system",
            "content": [
                {
                    "type": "text",
                    "text": long_text_content,
                    # cache_control can only be added to message content, not to tools
                    "cache_control": {"type": "ephemeral"},
                }
            ],
        },
        {"role": "user", "content": user_input},
    ]
    completion = client.chat.completions.create(
        model="qwen3.7-max",
        messages=messages,
        tools=tools,
        extra_body={"enable_thinking": False},
    )
    usage = completion.usage
    print(f"  Created: {usage.prompt_tokens_details.cache_creation_input_tokens}, "
          f"Hit: {usage.prompt_tokens_details.cached_tokens}")
    tool_calls = completion.choices[0].message.tool_calls
    if tool_calls:
        print(f"  Tools called: {[t.function.name for t in tool_calls]}")
    return completion

# First request: creates cache (includes tool definitions)
print("Q1: What's the weather in Beijing today?")
ask("What's the weather in Beijing today?")

# Second request: hits cache
print("\nQ2: Find flights from Shanghai to Beijing tomorrow")
ask("Find flights from Shanghai to Beijing tomorrow")

Expected output:

Q1: What's the weather in Beijing today?
  Created: 1995, Hit: 0
  Tools called: ['get_weather']

Q2: Find flights from Shanghai to Beijing tomorrow
  Created: 0, Hit: 1995
  Tools called: ['search_flights']