DeepSeek-V3 is a strong Mixture-of-Experts (MoE) language model that delivers exceptional performance and speed, surpassing its predecessors. With its advanced capabilities, DeepSeek-V3 ranks among the top open-source models and matches the performance of leading closed-source models.

In this notebook, we will demonstrate how to use the DeepSeek-V3 model for AgentChat within the AG2 framework.

​
Requirements

To get started, ensure you meet the following requirements:

  1. Install the AG2 package:

    • Run the following command to install the AG2 package:
    pip install ag2

  2. Obtain a DeepSeek API Key:

    • Sign up for a DeepSeek account here. to generate your API key.
    • Refer to the official documentation for more information about obtaining and using the API key.

​
Set the config for the DeepSeek API

Next, you need to configure the OAI_CONFIG_LIST to grant the agent access to the DeepSeek-V3 model.

Here’s an example configuration:

[
    {
        "model": "deepseek-chat",
        "base_url": "https://api.deepseek.com/v1",
        "api_key": "your DeepSeek Key goes here",
        "api_type": "deepseek",
        "tags": ["deepseek"]
    }
]

​
API parameters

You can include additional parameters in your configuration that are supported by the DeepSeek API. For a full list of parameters and their default values, refer to the DeepSeek API documentation.

Some examples of such parameters include:

  • temperature (number 0..1)
  • top_p (number 0..1)
  • max_tokens (null, integer >= 0)

Example Configuration:

[

    {
        "model": "deepseek-chat",
        "base_url": "https://api.deepseek.com/v1",
        "api_key": "your DeepSeek Key goes here",
        "api_type": "deepseek",
        "temperature": 0.5,
        "top_p": 0.2, # Note: It is recommended to set temperature or top_p but not both.
        "max_tokens": 10000,
        "tags": ["deepseek"]
    }
]

​
Two-Agent Coding Example

In this example, we run a two-agent chat with an AssistantAgent (primarily a coding agent) to generate code to count the number of prime numbers between 1 and 10,000 and then it will be executed.

We’ll use DeepSeek-V3 in our example.

import os

config_list = [
    {
        "model": "deepseek-chat",
        "base_url": "https://api.deepseek.com/v1",
        "api_key": os.environ.get("DEEPSEEK_API_KEY"),
        "api_type": "deepseek",
        "tags": ["deepseek"],
    }
]

from pathlib import Path

from autogen import AssistantAgent, UserProxyAgent
from autogen.coding import LocalCommandLineCodeExecutor

# Setting up the code executor
workdir = Path("coding")
workdir.mkdir(exist_ok=True)
code_executor = LocalCommandLineCodeExecutor(work_dir=workdir)

# Setting up the agents

# The UserProxyAgent will execute the code that the AssistantAgent provides
user_proxy_agent = UserProxyAgent(
    name="User",
    code_execution_config={"executor": code_executor},
    is_termination_msg=lambda msg: "FINISH" in msg.get("content"),
)

system_message = """You are a helpful AI assistant who writes code and the user executes it.
Solve tasks using your coding and language skills.
In the following cases, suggest python code (in a python coding block) for the user to execute.
Solve the task step by step if you need to. If a plan is not provided, explain your plan first. Be clear which step uses code, and which step uses your language skill.
When using code, you must indicate the script type in the code block. The user cannot provide any other feedback or perform any other action beyond executing the code you suggest. The user can't modify your code. So do not suggest incomplete code which requires users to modify. Don't use a code block if it's not intended to be executed by the user.
Don't include multiple code blocks in one response. Do not ask users to copy and paste the result. Instead, use 'print' function for the output when relevant. Check the execution result returned by the user.
If the result indicates there is an error, fix the error and output the code again. Suggest the full code instead of partial code or code changes. If the error can't be fixed or if the task is not solved even after the code is executed successfully, analyze the problem, revisit your assumption, collect additional info you need, and think of a different approach to try.
When you find an answer, verify the answer carefully. Include verifiable evidence in your response if possible.
IMPORTANT: Wait for the user to execute your code and then you can reply with the word "FINISH". DO NOT OUTPUT "FINISH" after your code block."""

# The AssistantAgent, using DeepSeek-V3 model, will take the coding request and return code
assistant_agent = AssistantAgent(
    name="DeepSeek_Assistant",
    system_message=system_message,
    llm_config={"config_list": config_list},
)

# Start the chat, with the UserProxyAgent asking the AssistantAgent the message
chat_result = user_proxy_agent.initiate_chat(
    assistant_agent,
    message="Provide code to count the number of prime numbers from 1 to 10000.",
)

User (to DeepSeek_Assistant):

Provide code to count the number of prime numbers from 1 to 10000.

--------------------------------------------------------------------------------
DeepSeek_Assistant (to User):

To count the number of prime numbers from 1 to 10000, we can use the Sieve of Eratosthenes algorithm, which is an efficient way to find all primes less than a given number. Here's the Python code to achieve this:

'''python
def count_primes(n):
    # Create a boolean array "prime[0..n]" and initialize
    # all entries it as true. A value in prime[i] will
    # finally be false if i is Not a prime, else true.
    prime = [True for _ in range(n+1)]
    p = 2
    while (p * p <= n):
        # If prime[p] is not changed, then it is a prime
        if prime[p] == True:
            # Updating all multiples of p to not prime
            for i in range(p * p, n+1, p):
                prime[i] = False
        p += 1

    # Counting prime numbers
    prime_count = 0
    for p in range(2, n+1):
        if prime[p]:
            prime_count += 1

    return prime_count

# Count the number of prime numbers from 1 to 10000
n = 10000
print(f"Number of prime numbers from 1 to {n}: {count_primes(n)}")
'''

This code will output the number of prime numbers between 1 and 10000. The Sieve of Eratosthenes algorithm works by iteratively marking the multiples of each prime number starting from 2, and then counting the remaining unmarked numbers as primes.

--------------------------------------------------------------------------------

>>>>>>>> NO HUMAN INPUT RECEIVED.

>>>>>>>> USING AUTO REPLY...

>>>>>>>> EXECUTING CODE BLOCK (inferred language is python)...
User (to DeepSeek_Assistant):

exitcode: 0 (execution succeeded)
Code output: Number of prime numbers from 1 to 10000: 1229


--------------------------------------------------------------------------------
DeepSeek_Assistant (to User):

The code executed successfully and determined that there are **1229 prime numbers** between 1 and 10000. This result is correct and aligns with known mathematical results.

If you have any further questions or need assistance with another task, feel free to ask!

FINISH.

--------------------------------------------------------------------------------

>>>>>>>> NO HUMAN INPUT RECEIVED.

​
Tool Call Example

In this example, instead of writing code, we will show how we can use DeepSeek-V3 model to perform parallel tool calling, where it recommends calling more than one tool at a time.

We’ll use a simple travel agent assistant program where we have a couple of tools for weather and currency conversion.

We start by importing libraries and setting up our configuration to use DeepSeek-V3 model.

import json
import os
from typing import Literal

from typing_extensions import Annotated

import autogen

config_list = [
    {
        "model": "deepseek-chat",
        "base_url": "https://api.deepseek.com/v1",
        "api_key": os.environ.get("DEEPSEEK_API_KEY"),
        "api_type": "deepseek",
        "tags": ["deepseek"],
    }
]

# Create the agent for tool calling
chatbot = autogen.AssistantAgent(
    name="chatbot",
    system_message="""For currency exchange and weather forecasting tasks,
        only use the functions you have been provided with.
        Output 'HAVE FUN!' when an answer has been provided.""",
    llm_config={"config_list": config_list},
)

# Note that we have changed the termination string to be "HAVE FUN!"
user_proxy = autogen.UserProxyAgent(
    name="user_proxy",
    is_termination_msg=lambda x: x.get("content", "") and "HAVE FUN!" in x.get("content", ""),
    human_input_mode="NEVER",
    max_consecutive_auto_reply=1,
)

# Create the two functions, annotating them so that those descriptions can be passed through to the LLM.
# We associate them with the agents using `register_for_execution` for the user_proxy so it can execute the function and `register_for_llm` for the chatbot (powered by the LLM) so it can pass the function definitions to the LLM.

# Currency Exchange function

CurrencySymbol = Literal["USD", "EUR"]

# Define our function that we expect to call


def exchange_rate(base_currency: CurrencySymbol, quote_currency: CurrencySymbol) -> float:
    if base_currency == quote_currency:
        return 1.0
    elif base_currency == "USD" and quote_currency == "EUR":
        return 1 / 1.1
    elif base_currency == "EUR" and quote_currency == "USD":
        return 1.1
    else:
        raise ValueError(f"Unknown currencies {base_currency}, {quote_currency}")


# Register the function with the agent


@user_proxy.register_for_execution()
@chatbot.register_for_llm(description="Currency exchange calculator.")
def currency_calculator(
    base_amount: Annotated[float, "Amount of currency in base_currency"],
    base_currency: Annotated[CurrencySymbol, "Base currency"] = "USD",
    quote_currency: Annotated[CurrencySymbol, "Quote currency"] = "EUR",
) -> str:
    quote_amount = exchange_rate(base_currency, quote_currency) * base_amount
    return f"{format(quote_amount, '.2f')} {quote_currency}"


# Weather function


# Example function to make available to model
def get_current_weather(location, unit="fahrenheit"):
    """Get the weather for some location"""
    if "chicago" in location.lower():
        return json.dumps({"location": "Chicago", "temperature": "13", "unit": unit})
    elif "san francisco" in location.lower():
        return json.dumps({"location": "San Francisco", "temperature": "55", "unit": unit})
    elif "new york" in location.lower():
        return json.dumps({"location": "New York", "temperature": "11", "unit": unit})
    else:
        return json.dumps({"location": location, "temperature": "unknown"})


# Register the function with the agent


@user_proxy.register_for_execution()
@chatbot.register_for_llm(description="Weather forecast for US cities.")
def weather_forecast(
    location: Annotated[str, "City name"],
) -> str:
    weather_details = get_current_weather(location=location)
    weather = json.loads(weather_details)
    return f"{weather['location']} will be {weather['temperature']} degrees {weather['unit']}"

# start the conversation
res = user_proxy.initiate_chat(
    chatbot,
    message="What's the weather in New York and can you tell me how much is 123.45 EUR in USD so I can spend it on my holiday? Throw a few holiday tips in as well.",
)

user_proxy (to chatbot):

What's the weather in New York and can you tell me how much is 123.45 EUR in USD so I can spend it on my holiday? Throw a few holiday tips in as well.

--------------------------------------------------------------------------------
chatbot (to user_proxy):


***** Suggested tool call (call_0_5c4b1925-5eec-4190-9887-fecf4dcc4c07): weather_forecast *****
Arguments:
{"location": "New York"}
***********************************************************************************************
***** Suggested tool call (call_1_f75a7688-2c44-4a81-b8dd-c32b71f5750d): currency_calculator *****
Arguments:
{"base_amount": 123.45, "base_currency": "EUR", "quote_currency": "USD"}
**************************************************************************************************

--------------------------------------------------------------------------------

>>>>>>>> EXECUTING FUNCTION weather_forecast...
Call ID: call_0_5c4b1925-5eec-4190-9887-fecf4dcc4c07
Input arguments: {'location': 'New York'}

>>>>>>>> EXECUTING FUNCTION currency_calculator...
Call ID: call_1_f75a7688-2c44-4a81-b8dd-c32b71f5750d
Input arguments: {'base_amount': 123.45, 'base_currency': 'EUR', 'quote_currency': 'USD'}
user_proxy (to chatbot):

***** Response from calling tool (call_0_5c4b1925-5eec-4190-9887-fecf4dcc4c07) *****
New York will be 11 degrees fahrenheit
************************************************************************************

--------------------------------------------------------------------------------
***** Response from calling tool (call_1_f75a7688-2c44-4a81-b8dd-c32b71f5750d) *****
135.80 USD
************************************************************************************

--------------------------------------------------------------------------------
chatbot (to user_proxy):

Here's the weather in New York and the currency conversion:

- **Weather in New York**: 11°F
- **Currency Conversion**: 123.45 EUR is approximately 135.80 USD

### Holiday Tips for New York:
1. **Visit Central Park**: A beautiful place to relax and enjoy nature in the middle of the city.
2. **Explore Times Square**: Experience the vibrant lights and energy of this iconic location.
3. **Statue of Liberty**: Take a ferry ride to see this historic monument up close.
4. **Broadway Shows**: Catch a world-class performance in the theater district.
5. **Museums**: Don't miss the Metropolitan Museum of Art and the American Museum of Natural History.

HAVE FUN!

--------------------------------------------------------------------------------

We can see that DeepSeek-V3 recommended invoking both tools with the appropriate parameters. The user_proxy executed these calls, and the results were passed back to DeepSeek-V3 for interpretation and response.

CohereGoogle Gemini