What is Hermes Agent?

Hermes Agent is an open-source AI agent developed by Nous Research, engineered to execute complex tasks, retain acquired knowledge, and evolve its capabilities over time. Beyond simple response generation, it leverages tools such as terminals, browsers, file systems, web search, and code execution environments to carry out substantive work.

What is an AI agent?

An AI agent extends beyond basic text generation. When provided with a goal, it autonomously determines the necessary actions, utilizes available tools, observes outcomes, and iteratively progresses toward the objective.

While many AI agents can already employ tools to complete multi-step tasks, the critical differentiator lies in post-task behavior. Some agents treat every interaction as an isolated problem. In contrast, Hermes is designed to retain valuable knowledge and procedural insights, enabling it to become increasingly effective over time. To understand how this applies to running agents on DaDesktop, see Agentic AI.

What makes Hermes different?

Hermes features an integrated learning loop. It generates skills from past experiences, refines them during use, remembers critical information, and retrieves data from previous sessions when context from earlier work is required.

Why use Hermes Agent?

Hermes is particularly valuable for tasks involving multiple steps, tools, or applications. Rather than manually switching between AI chat interfaces, terminals, browsers, and file managers, you can assign the overarching task to Hermes and allow it to orchestrate the workflow.

• Execute complete tasks: Assign a high-level objective to Hermes instead of requesting individual instructions.
• Maintain long-term context: Preserve relevant information across separate sessions.
• Create reusable workflows: Store effective strategies as skills to avoid re-solving identical problems from scratch.
• Automate recurring work: Schedule tasks to execute automatically in the background.
• Delegate complex work: Utilize separate subagents to handle different components of a larger task.
• Select your model: Connect Hermes to local models or supported cloud model providers.

What Hermes can do

• Software development: Analyze code, debug issues, implement changes, run tests, and manage repositories.
• Research: Search the web, review sources, gather data, and structure findings.
• Automation: Execute scripts and schedule recurring tasks, including reports, monitoring, and other routines.
• System administration: Operate via the terminal, install software, configure systems, and resolve technical issues.
• Browser interactions: Navigate websites, interact with web pages, and extract information.
• File management: Read, create, modify, search, and organize files.
• Tool integrations: Connect additional services and tools via MCP.

How Hermes improves with use

Consider a complex task requiring multiple steps. During execution, Hermes may identify an efficient workflow. Instead of discarding this knowledge upon task completion, it can save the workflow as a skill.

When a similar task arises, Hermes can leverage this skill as a foundation. Should the workflow require refinement, the skill can be updated accordingly. Its memory system also retains key facts and lessons learned from prior operations.

Memory and skills serve distinct functions. Memory archives useful information and context, while skills encapsulate procedural knowledge. Together, they enable Hermes to accumulate a comprehensive understanding of both what it knows and how to execute recurring tasks.

Automate background work

Hermes includes a built-in scheduler for recurring tasks. You can define schedules using natural language, allowing Hermes to perform work autonomously without requiring an active session initiation.

This capability is ideal for reports, monitoring, backups, research, and other routine operations. Scheduled tasks can maintain continuity between executions, enabling them to build upon previous results.

Run local or cloud models

Hermes is not restricted to a single LLM. You can connect it to local models or supported cloud providers, switching models without altering the agent's tools, memory, or workflows.

Utilizing a local model grants greater control over the model, hardware, and data processing location. It also allows Hermes to interact with your files and local resources without transmitting them to external model providers.

How to install Hermes

Hermes is compatible with Linux, macOS, Windows, and WSL2. You can install the desktop application or utilize the command-line interface independently of the desktop version.

On Linux, macOS, or WSL2, open a terminal and execute the following command:

curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash

On Windows, run the installer from PowerShell using:

irm https://hermes-agent.nousresearch.com/install.ps1 | iex

Post-installation, launch the setup wizard to configure your model provider and other settings:

hermes setup

Consult the Hermes documentation for detailed installation options and configuration parameters.

Run Hermes on DaDesktop

Hermes requires an environment capable of executing its tools and storing data. DaDesktop offers a cloud desktop equipped with the CPU, RAM, storage, and GPU resources necessary to run Hermes alongside a local LLM.

This is particularly beneficial when the desired model requires more VRAM than your local hardware supports. Select a GPU with sufficient VRAM, launch a DaDesktop instance, install Hermes, and run the model locally within the same environment as the agent.

You can also keep the desktop active for scheduled tasks and long-running workflows, eliminating the need to leave your personal computer powered on.

Learn more about running local LLMs on DaDesktop.