Chat and everyday writing
Hermes Agent
Made by Hermes
An open, MIT-licensed agent you point at a task — free software, you pay your own model.
What it is
Hermes Agent isn't a chatbot. It's an open agent: you describe a job, it plans steps and carries them out using whichever model's API key you give it. The software costs nothing and the license lets you do what you like with it.
It belongs here mainly as a way to understand what people mean by an 'agent' without paying a subscription to find out. Expect rougher edges than the commercial tools, and expect to read some documentation.
Because it runs on your own model key, your running cost is whatever that provider charges — Hermes takes nothing.
Published price
- SoftwareMIT license, run it yourself$0
- Model useNo fee to HermesWhatever your own API key costs
- Managed hosting — 7-day trialIncludes built-in model access, 500 conversations every 5 hours$12 one-off
- Managed hosting — monthlyThird-party hosts run this; Hostinger's managed plan starts at $5.99/month and renews at $11.99/monthFrom about $6/month
- Your own server instead$4–$25/month VPS plus your model spend
The project itself sells nothing; the hosted figures come from the third-party hosts that package it, and the official hosting page prints only the trial price.
Source: the Hermes Agent site · Last checked 17 September 2026
Benchmarks
The published figures for Hermes Agent, copied exactly as their sources give them and checked 17 September 2026. Each one names the test, who ran it and when, and says when the number comes from the maker rather than an independent tester.
- MMLU-Pro82.9%
hard multiple-choice questions across professional subjects · tested: Hermes 4 405B
Benchmark List, citing the Hermes 4 report · 13 August 2025 report · maker's own figures
- GPQA Diamond72.7%
graduate-level science questions · tested: Hermes 4 405B
Nous Research technical report · 13 August 2025 · maker's own figures
- Artificial Analysis Intelligence Index9
combined score across ten reasoning and knowledge tests · tested: Hermes 4 405B
Artificial Analysis · checked September 2026 · independent
Far below the hosted flagships — this is the price of running it yourself.
The agent itself has never been benchmarked. These are scores for Hermes 4 405B, the model it runs on, and the headline numbers come from Nous Research's own technical report.
Figures copied from the sources above, checked 17 September 2026.
Where to read today's numbers
Scores move, so these are the boards that keep them current.
- SWE-bench leaderboardIndependent
Measures whether an agent can fix real GitHub issues. The standard reference for coding agents.
- Agent evaluations on Vals AIIndependent
Use these to judge the underlying model you plug in; the agent wrapper is not separately scored.
How it performs, in my experience
An open agent you point at a task rather than a chatbot; capable but rougher around the edges than the commercial assistants.
Free value
5 / 5
How much you get without paying.
Performance
3 / 5
How well it does its main job.
Range of uses
3 / 5
How many different jobs it suits.
Free and open, rougher than the commercial tools — good for learning agents. These scores are my own judgment, not a measurement — the benchmark links above are the independent version.
Prompting it from this site
Can't be run from here
It's an agent that works across a whole project, not a single prompt. Run it on your own machine — the setup steps on its page show how.
Use it for
- Automating a repeated multi-step task
- Learning how agents actually work
Running it yourself
Runs on your own machine
Hermes is open-weight, and the smaller sizes run on an ordinary desktop or a well-specified laptop. This is the easiest local model on this list.
What you need first
- macOS, Linux or Windows
- Roughly 8GB of free memory for a small size, more for the larger ones
- Ollama, which bundles everything else
Step by step
1.Install Ollama
curl -fsSL https://ollama.com/install.sh | sh2.Pull and run the model
ollama run hermes33.Or serve it to your own code on the OpenAI-compatible port
ollama serve # then point your client at http://localhost:11434/v1
On Windows and macOS, download the Ollama installer from ollama.com rather than using the shell script.