FrameworkDeep Agents with DEVUP AI
Run Deep Agents on DEVUP AI models. Connect ChatOpenAI to DEVUP AI and give the agent your own tools alongside its built-in file tools.
System Architecture
How Deep Agents Connects to DEVUP AI
1. Client
Deep Agents
Your Python agent
2. Gateway
DEVUP OpenAI-compatible API
POST /v1/chat/completions
3. Inference
Model
DeepSeek-V4-Pro
4. Response
Response
Tool calls, files and answers
Configuration
Quick Setup
bash
pip install deepagents langchain-openaipython
import os
from deepagents import create_deep_agent
from langchain_openai import ChatOpenAI
model = ChatOpenAI(
model="deepseek-ai/DeepSeek-V4-Pro",
base_url="https://api.devupai.com/v1",
api_key=os.environ["DEVUP_API_KEY"],
)
def multiply(a: int, b: int) -> int:
"""Multiply two integers."""
return a * b
agent = create_deep_agent(
model=model,
tools=[multiply],
system_prompt="You are a precise assistant. Use the multiply tool for arithmetic.",
)
result = agent.invoke({"messages": [{"role": "user", "content": "What is 12 multiplied by 34? Reply with the number only."}]})
tool_calls = [call["name"] for message in result["messages"] for call in getattr(message, "tool_calls", [])]
print(tool_calls, result["messages"][-1].content)Virtual Filesystem
Files in the Agent State
python
result = agent.invoke({"messages": [{"role": "user", "content": "Use write_file to save the text Salam DEVUP to /notes.md, then reply with done."}]})
print(sorted(result["files"]))Capabilities
Supported Capabilities
Custom tools
Pass Python functions as tools; the agent calls them through tool calling.
Virtual filesystem
Built-in file tools write and edit files that are kept in the agent state.
Built on LangGraph
create_deep_agent returns a LangGraph graph that you run with invoke.
Local DZD billing
Metered billing in Algerian Dinar with BaridiMob, Edahabia, and CIB local payment support.
