Alibaba open-sources CoPaw as a local-first alternative to cloud AI agents

ByIvan Jenic

Alibaba’s AgentScope team just released a new open-source agent workstation. CoPaw lets developers run multi-channel AI assistants locally with persistent memory.

Alibaba Copaw
Image: GitHub/alibaba

Alibaba’s AgentScope team officially released CoPaw. This open-source framework acts as a personal agent workstation. It enables developers to deploy multi-channel AI assistants entirely on local hardware.

Key Takeaways:

  • The core utility: The framework functions as a self-hosted workstation. It directly rivals self-hosted task automation tools like OpenClaw. Users can orchestrate multiple agents to collaborate on research and productivity tasks.
  • The memory system: CoPaw solves a major pain point for autonomous agents. It uses a persistent memory system called ReMe. Agents retain context and instructions across long-running workflows and multiple days.
  • The deployment: The system is built for local-first execution. It runs through Ollama and supports models like Qwen 3.5 and Llama variants. Developers can also deploy it via a one-click script on Alibaba Cloud ECS.
  • The integrations: The workstation connects to several applications simultaneously. Agents can manage tasks across Discord, Telegram, Slack, WhatsApp, and email without dropping context.
  • The hardware barrier: Running capable local models requires significant computing power. This hardware demand limits accessibility for users without dedicated GPUs.
  • The availability: The project is fully open-source under the Apache 2.0 license. Developers can access the code on GitHub or view the documentation on the official website.

The Bottom Line: CoPaw provides the necessary infrastructure to run persistent, multi-channel AI agents without paying for proprietary cloud APIs.

RunPod
RunPod

If you need on-demand GPUs for training, fine-tuning, inference, or running open-source models, give RunPod a try.

  • Available hardware: H100, H200, A100, L40S, RTX 4090, RTX 5090, and 30+ more
  • Cost: significantly cheaper than AWS or GCP, billed per second, no contracts
  • Setup: spins up in under a minute, 30+ regions worldwide
Try RunPod →
Affiliate disclosure: We may earn a commission if you sign up via our link, at no extra cost to you.
Efficienist Newsletter

Get the core business tech news delivered straight to your inbox. We track AI, automation, SaaS, and cybersecurity so you don't have to.

Just read what you want, and be done with it.

Read Next