Here’s a smart, cheap setup. Pairing the Hermes agent with Qwen 3.8 is one of the best cheap-and-powerful setups going. Qwen 3.8 is Alibaba’s latest open-source model, it’s excellent at agentic and coding tasks, and it plugs straight into Hermes as the brain.
Here’s why the combo works and how to set it up.
Last updated: July 2026.
Key takeaways
- Qwen 3.8 is a top Chinese open-source model — strong at building and automating, and cheap to run.
- Hermes is the hands, Qwen 3.8 is the brain — it plugs straight in.
- It’s one of my favourite open-source picks, alongside GLM 5.2.
- The ready-made setup is in the AI Profit Boardroom.
What Is Qwen 3.8?
Qwen 3.8 is the newest model in Alibaba’s Qwen family — an open-source model built for real work, not just chat. It’s genuinely capable at coding, building and automating, which is exactly what you want behind an AI agent. Among Chinese open-source models, Qwen is one of the two I reach for most (the other being GLM 5.2).
Because it’s open source, it’s cheap — and often free — to run, which makes it a smart brain for an always-on agent.
Why Use Qwen 3.8 With Hermes?
Hermes is a free, open-source agent, but it’s only as good as the model driving it. Qwen 3.8 is a strong choice: it follows agentic workflows well, it’s good at building things, and it won’t drain your budget the way a frontier API can.
Think of it as brain and hands — Qwen 3.8 does the thinking, Hermes does the doing (files, tools, browsing, automation). Together they build and automate genuinely useful things.
How to Connect Qwen 3.8 to Hermes
- Open your terminal and type
Hermes model. - Pick Qwen in the model dropdown (via your provider — OpenRouter, or a local runner like Ollama once it’s available).
- Log in / add your key and set it as the brain.
- Or use the Hermes web UI — go to Models and select Qwen there.
New to Hermes? Start with my Hermes install guide, then swap in Qwen 3.8.
Qwen 3.8 vs Other Hermes Brains
| Model | Best for | Cost |
|---|---|---|
| Qwen 3.8 | Building & automating, agentic work | Cheap / often free |
| GLM 5.2 | Fast, high-volume everyday tasks | Very cheap |
| Kimi K3 | Frontier-grade quality | Cheap for its level |
| Gemma 4 | Free local / offline | Free |
| Opus 4.8 / Fable 5 | Top-quality, high-stakes | Premium |
Match the model to the job — see my best model for Hermes agent guide for the full rundown.
What You Can Build
With Qwen 3.8 as the brain, Hermes can build websites and tools, run automations, research and organise information, and pair with systems like NotebookLM for content workflows. It’s a build-and-automate machine that costs a fraction of a frontier model.
For a similar cheap-and-powerful pairing, see my Kimi K3 + Hermes guide.
Is Qwen 3.8 Good for Coding?
Yes — coding is one of Qwen’s strengths. It writes and edits code well, which makes it a capable coding brain inside Hermes for building sites, tools and automations. It won’t out-polish a top frontier model on the hardest tasks, but for the price it’s excellent, and for everyday building it’s more than enough.
Qwen 3.8 vs Kimi K3 in Hermes
Both are strong open-source brains for Hermes. Kimi K3 pushes closer to frontier quality; Qwen 3.8 is a superb cheap all-rounder for building and automating. If cost matters most, use Qwen; if you want the highest quality from an open model, Kimi K3 — see my Kimi K3 tests.
Running Qwen 3.8 Free vs Paid
You’ve got options on cost. Run Qwen 3.8 through a free or cheap API and running costs stay tiny; run it locally (once local support lands) and it’s effectively free after setup. Either way it’s far cheaper than a frontier API for the same agentic work — which is the whole point of an open-source brain. For the free angle, see my best free AI model guide.
Get the Setup Ready-Made
The full Hermes + Qwen setup — models, memory and workflows wired together — is inside the AI Profit Boardroom Agent OS.
New here? Start free with my AI course and community (plus 1,000+ AI agents), or grab a free strategy session.
FAQ
Is Qwen 3.8 good for Hermes agent?
Yes — it’s a strong open-source model for agentic and coding work, cheap to run, and it plugs straight into Hermes as the brain.
How do I use Qwen 3.8 with Hermes?
Type Hermes model in your terminal, pick Qwen via your provider (or the Hermes web UI Models page), add your key and set it as the brain.
Is Qwen 3.8 free?
It’s open source and often available free via APIs, and cheap to run otherwise — far cheaper than a frontier model for the same agentic tasks.
Qwen 3.8 vs GLM 5.2 for Hermes?
Both are excellent cheap open-source brains. Qwen is great for building and automating; GLM 5.2 is fast for high-volume everyday work. Many people use both.
Do I need a powerful computer?
Not if you run Qwen via an API. You only need strong hardware if you run it locally; otherwise any modern machine is fine.
The Bottom Line
Hermes with Qwen 3.8 delivers agentic power on a budget. The full setup is in the AI Profit Boardroom.