OpenClaw

This Ubuntu 24.04-based image includes pre-installed OpenClaw (version 2026.2.6-3) — a framework for creating autonomous AI agents capable of executing real-world tasks: from data analysis and code generation to API interactions and infrastructure management.

Unlike GPU-dependent solutions, OpenClaw runs efficiently on a CPU. This is because its core workload involves orchestration, decision-making logic, and interfacing with external models, rather than local inference.

The image also features NGINX 1.24+ pre-configured as a reverse proxy, granting you immediate access to the OpenClaw Control web interface via your browser right after deployment.

Why OpenClaw has become so popular

Unlike most AI agents that are limited to predefined templates or work with only a single model, OpenClaw is built as a universal platform for autonomous agents. Its main strength lies in its flexibility and open architecture:

• Support for any provider: You can simultaneously connect public APIs and private endpoints — for example, LLMs deployed in the cloud or on a local network.
• Skills system: The agent doesn't just answer queries — it plans a sequence of actions, calls the necessary tools, analyzes the results, and adapts to errors.
• Full data control: Since inference can be routed through your own endpoints, your source code, internal documents, and business logic never leave your infrastructure.
• Seamless integration: Thanks to OpenAPI compatibility and support for standard formats, OpenClaw easily integrates into existing CI/CD pipelines, monitoring systems, and internal chat interfaces.

By hosting a cloud server on immers.cloud, you get a ready-to-use environment for running OpenClaw — without the hassle of dealing with drivers, dependencies, and hardware configuration. This is a perfect solution not only for developers who need a stable and high-performance stack for experimenting with AI agents, but also for those who want to build a custom AI assistant tailored to their specific tasks.

Minimum Resource Requirements

For stable operation of OpenClaw, a minimum of 4 GB of RAM is recommended. This is sufficient to prevent out-of-memory errors even under moderate workloads.

If you plan to run multiple agents simultaneously, enable detailed logging, or connect models with local inference (even on a CPU), we highly recommend choosing a configuration with 4 GB of RAM or more. This will ensure smooth performance and prevent slowdowns during peak loads.

Kernel version:
Linux 6.8.0-100-generic #100-Ubuntu SMP PREEMPT_DYNAMIC Tue Jan 13 16:40:06 UTC 2026 x86_64 x86_64 x86_64 GNU/Linux

Operating System version:
No LSB modules are available.
Distributor ID: Ubuntu
Description: Ubuntu 24.04.3 LTS
Release: 24.04
Codename: noble