Deploy Your Own Custom AI Assistant on Alibaba Cloud in 5 Minutes
This guide walks you through setting up the open‑source CoPaw personal AI agent on Alibaba Cloud PAI‑EAS, covering why the platform is advantageous, the quick 5‑minute deployment steps, public network configuration, model integration, and how to start interacting with your private assistant.
Why Choose CoPaw on PAI‑EAS?
CoPaw is a locally‑first, self‑hosted personal AI agent framework released by Alibaba Cloud AgentScope. It offers a fully integrated "dedicated model + dedicated agent" experience, giving you high model freedom, fine‑grained cost control, and seamless integration with popular office channels such as DingTalk and Feishu.
Model freedom: Switch between Alibaba Cloud Baileian model resources and custom model services on PAI‑EAS.
Cost control: Deep integration with Aliyun Coding Plan keeps inference costs transparent.
All‑scenario assistant: Native support for DingTalk, Feishu, etc., embedding the agent into daily workflows.
5‑Minute Rapid Deployment Guide
Step 1 – Deploy the EAS Service
Log in to the PAI console (choose region and workspace) and click Enter EAS .
In the Inference Service tab, click Deploy Service → Custom Model Deployment → Custom Deploy .
Configure the following key parameters:
Service Name: e.g., copaw_demo Deployment Method: Image deployment, enable Web Application .
Image Configuration: Choose the official copaw image, latest version.
Storage Mount:
Uri: oss://examplebucket/copaw/ Mount Path: /mnt/data/ Run Command: /start.sh Port: 8088 Resource Type: Public resource.
Deploy Resource: CPU type ecs.c7a.large (adjust as needed).
VPC, Subnet, Security Group: Create and configure the appropriate VPC, switch, and security group.
Click Deploy and wait for the service to become available.
Step 2 – Configure Public Network Access
To expose the EAS service to the internet, set up a NAT gateway with an Elastic IP (EIP) and a SNAT entry.
Create a public NAT gateway and bind an EIP (pay‑as‑you‑go).
Select the same region as the EAS service and choose the VPC created in Step 1.
Configure a SNAT entry with VPC scope and associate the previously bound EIP.
Step 3 – Configure the Model and Start a Conversation
After the service is running, open the WebUI via the "Call/Log/Monitor" button in the console.
In the WebUI, go to Settings → Model and choose a provider and LLM region. Three configuration options are provided:
(1) Aliyun Coding Plan: Subscribe on the Coding Plan page to obtain a dedicated API Key.
(2) Add an EAS Model Service:
Enter the Model Gallery, select a model (e.g., Qwen3.5-397B-A17B), and deploy it.
After deployment, note the service endpoint and token.
Add a new provider in CoPaw, choose OpenAI compatible , set the URL to {endpoint}/v1 and the API Key to the token.
Add the model to the provider configuration.
(3) DashScope: Add the Baileian API Key, then select additional models from the DashScope model list.
Finally, you can start chatting with your private CoPaw assistant.
Optional – Channel Integration
Follow the channel configuration guide to connect CoPaw with DingTalk or Feishu, enabling the assistant to operate directly within those collaboration tools.
🎉 Open the chat window and issue your first task to the CoPaw personal assistant!
Alibaba Cloud Big Data AI Platform
The Alibaba Cloud Big Data AI Platform builds on Alibaba’s leading cloud infrastructure, big‑data and AI engineering capabilities, scenario algorithms, and extensive industry experience to offer enterprises and developers a one‑stop, cloud‑native big‑data and AI capability suite. It boosts AI development efficiency, enables large‑scale AI deployment across industries, and drives business value.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.