Multi-Agent Framework

AutoGen with DEVUP AI

Build next-generation multi-agent conversational systems using Microsoft AutoGen and DEVUP AI. By configuring your agent config_list with our base URL, your agents collaborate, generate code, execute solutions, and iterate autonomously across 200+ catalog models.

Get API Key AutoGen Documentation

Under the hood: AutoGen communicates with DEVUP AI via standard OpenAI-compatible client configurations by specifying base_url: "https://api.devupai.com/v1" in config_list.

OpenAI Gateway Specs

Architecture Flow

How AutoGen Connects to DEVUP AI

AutoGen dispatches agent messages through an internal OpenAI wrapper that targets your configured endpoint. Responses are parsed by the agent runtime to drive multi-turn reasoning and local code execution.

1. Agent Layer

AutoGen Agents

Assistant & UserProxy

2. Orchestration

Conversation Loop

Turn routing & execution

3. Gateway

DEVUP AI /v1

https://api.devupai.com/v1

4. Execution

Target Model

DeepSeek-V4-Pro, Qwen, etc.

Installation

Install Packages

Install the Microsoft AutoGen package using pip:

bash

pip install pyautogen

Configuration

Configure Agent Endpoint

Define a config_list entry specifying the target model, DEVUP AI gateway URL, and your API key:

python

import os
import autogen

config_list = [
    {
        "model": "deepseek-ai/DeepSeek-V4-Pro",
        "base_url": "https://api.devupai.com/v1",
        "api_key": os.environ["DEVUP_API_KEY"],
    }
]

llm_config = {"config_list": config_list, "seed": 42}

Multi-Agent Orchestration

Autonomous Problem Solving & Code Execution

Create an AssistantAgent paired with an execution-enabled UserProxyAgent to solve coding tasks collaboratively:

python

import autogen

config_list = [
    {
        "model": "deepseek-ai/DeepSeek-V4-Pro",
        "base_url": "https://api.devupai.com/v1",
        "api_key": "<your DEVUP API key here>",
    }
]

llm_config = {"config_list": config_list, "seed": 42}

assistant = autogen.AssistantAgent("assistant", llm_config=llm_config)
user_proxy = autogen.UserProxyAgent("user_proxy", code_execution_config={"work_dir": "coding"})
user_proxy.initiate_chat(assistant, message="What time is it right now?")

Observed Execution

Autonomous Turn-by-Turn Output

During execution, the assistant generates solution code, the user proxy detects and executes it locally, and the result is returned to conclude the interaction:

text

user_proxy (to assistant):

What time is it now?

--------------------------------------------------------------------------------
assistant (to user_proxy):

To get the current time, you can use the `datetime` module in Python...

```python
import datetime
current_time = datetime.datetime.now()
print(current_time.strftime("%I:%M %p"))
```

--------------------------------------------------------------------------------
user_proxy (to assistant):

exitcode: 0 (execution succeeded)
Code output:
02:20 PM

--------------------------------------------------------------------------------
assistant (to user_proxy):

The current time is 02:20 PM.

Capabilities Matrix

Verified AutoGen Features

Conversational Multi-Agent

Automated peer-to-peer and group chat conversations between distinct AI agents.

Two-agent conversations via initiate_chat()
Multi-agent group collaboration via GroupChat
Human-in-the-loop oversight modes (TERMINATE, ALWAYS, NEVER)

Code Generation & Tooling

Executable scripts generated by models and run within sandboxed environments.

Automated Python code detection and execution
Self-correcting code iterations when execution errors occur
Custom tool and function registration on agents

config_list Schema Parameters

Field	Type	Required	Description
base_url	str	Yes	Must be set to "https://api.devupai.com/v1"
api_key	str	Yes	Your DEVUP AI API key (sk-devup-...)
model	str	Yes	Target model identifier from the platform catalog
temperature	float	No	Sampling temperature (e.g. 0.2 for deterministic code)
seed	int	No	Random seed for reproducible conversation turns

Build Multi-Agent Workflows with AutoGen Today

Get your API key, configure AssistantAgent, and build collaborative AI agent networks with Algerian Dinar billing.

Get API Key View Documentation