Yes, you can use Nemotron 3.5 Super VL free — and there are several ways to do it: inside OpenCode, through OpenRouter’s free variant, or by downloading the open weights. In the video below I use it free in OpenCode and Hermes. Here’s every free route, the limits to expect, and how to make the most of it.
Key takeaways
- OpenCode: free in the model drop-down at the time of writing.
- OpenRouter:
nvidia/nemotron-3.5-super-vl:freecosts nothing (rate limited). - Open weights: free to download from Hugging Face, Ollama and LM Studio.
- Free routes have limits — keep a backup model.
Every way to use Nemotron 3.5 Super VL free
| Free route | How | Catch |
|---|---|---|
| OpenCode | Select it in the model drop-down | Free window may change |
| OpenRouter free variant | Model id nvidia/nemotron-3.5-super-vl:free | Rate limits |
| Hermes Agent | Use the OpenRouter free variant as your model | Same OpenRouter limits |
| Open weights | Download from Hugging Face, Ollama or LM Studio | Needs 70GB+ of memory |
Free vs paid on OpenRouter
OpenRouter lists two versions: the free variant (:free, about 1M context) and the paid one at $0.15 per million input tokens and $0.50 per million output tokens, with a 1,048,576-token context. The paid version is cheap and avoids free-tier rate limits, so it’s worth it for heavy use.
Want this working in your business, not just bookmarked? Using free open models in your AI agents is exactly what we build together inside the AI Profit Boardroom — 3,400+ members, four live calls a week, and a full one-hour DeepSeek Harness course in the classroom.
Prefer it mapped 1-on-1 first? Book a free strategy session and we’ll plan it for your exact situation.
Join the AI Profit Boardroom →Claim Your Free SEO Strategy Session →
Running it free on your own hardware
The weights are open under the OpenMDW-1.1 licence, which allows commercial use. But “free” locally needs serious hardware: the Ollama build is 80–90GB and LM Studio’s smallest is about 70GB, needing at least 70GB of RAM. Nvidia’s full-precision version is designed for multiple data-centre GPUs. For most people, the free cloud routes are the practical option.
Getting the most from the free version
- Use it for jobs that start with a screenshot, PDF or chart — its strength.
- Keep prompts focused — free tiers have limits.
- Set up a backup free model for when you hit them.
- Upgrade to the cheap paid OpenRouter version for big jobs.
Is the free version as good?
The free variant is the same model, just with usage limits. The quality is identical; you’re trading speed and volume for zero cost.
Let me be clear: this is not a frontier model. It’s not going to replace Claude Opus 5.5 for your hardest work. What it is: a free, open, fast model that can read screenshots, PDFs and charts and then act on them — which is exactly the job a lot of agent work starts with.
What to do with a free vision model
A free model that can see opens up jobs that used to need paid tools. Turn photos of receipts into a spreadsheet. Pull the numbers out of a PDF report. Read screenshots of competitor pricing pages. Summarise slide decks. Check a chart in a client report and explain what changed. Each of these used to mean manual copying or a paid OCR service; now an agent can do them for free.
How to avoid hitting the limits
Free tiers have rate limits, so batch similar jobs together rather than firing off dozens of tiny requests. Crop images to what matters so each request is smaller. Use the free version for testing and light daily work, and switch to the paid OpenRouter version — still only $0.15 per million input tokens — when you run a big batch. And always keep a second free model ready in your agent so a limit never stops your workflow.
Free today, maybe not forever
Free windows for new models often change once the launch period ends, so build your workflows so the model can be swapped. In Hermes, that’s as simple as keeping the model in its own profile. If the free OpenCode access or the free OpenRouter variant changes, you switch the profile and carry on.
Final thoughts
To sum up the free options: OpenCode is the quickest to try, OpenRouter’s free variant works in almost any agent including Hermes, and the open weights are free if you have the hardware. Start with the one that matches the tool you already use, and you can be testing Nvidia’s newest vision model in minutes without spending anything.
If you want help building workflows around free models like this one, that’s what we do on the weekly coaching calls inside the AI Profit Boardroom.
How I’d test it this week
Pick one visual job you currently do by hand — copying numbers from a report, sorting receipts, reading screenshots — and give it to the free version this week. If it saves you even fifteen minutes a week, it’s worth keeping in your stack for nothing.
Once that works, add a second job, and keep the cheap paid version as your fallback for busy days.
Free Nemotron 3.5 Super VL in Hermes, step by step
Create a new Hermes profile for it so it doesn’t change your main setup. Run hermes model, choose OpenRouter, and select nvidia/nemotron-3.5-super-vl:free. Send it a test screenshot and a question. If it works, turn your first visual job into a skill and schedule it. When you hit a limit, switch that profile to the paid version or another free model — your skills and memory stay exactly the same. That’s the whole setup, and it costs nothing to start.
The bottom line on Nemotron 3.5 Super VL free
Using Nemotron 3.5 Super VL free is easy: pick it in OpenCode or use OpenRouter’s free variant in Hermes. Start there, and move to the cheap paid version only when you need more.
FAQ: nemotron 3.5 super vl free
Is Nemotron 3.5 Super VL free?
Yes — in OpenCode at the time of writing, via OpenRouter’s free variant, and as open weights.
What is the free OpenRouter model id?
nvidia/nemotron-3.5-super-vl:free.
How much does the paid version cost?
$0.15 per million input tokens and $0.50 per million output tokens on OpenRouter.
Can I run it free on my PC?
Only with 70GB+ of memory; most people should use the free cloud routes.
Is the free version worse?
No — it’s the same model with usage limits.
Can I use Nemotron 3.5 Super VL free in Hermes Agent?
Yes — choose OpenRouter in hermes model and select nvidia/nemotron-3.5-super-vl:free. Keep it in its own profile and switch to the paid version or another free model if you hit rate limits.
Two ways I can help from here. Join the AI Profit Boardroom — it’s where your AI agent model setup gets built with 3,400+ members doing the same.
Or grab a free strategy session and bring your questions.
Join the AI Profit Boardroom →Claim Your Free SEO Strategy Session →
About Julian Goldie
I’m Julian Goldie — founder of Goldie Agency, best-selling author, and an AI educator with 428K+ YouTube subscribers. I’ve spent 10+ years in SEO and online business and I test every AI agent on my own work before I recommend it. Join the AI Profit Boardroom for the daily builds, or book a free strategy session to talk through yours.
Related reading
Last updated October 2026. This page is a living guide to nemotron 3.5 super vl free — the facts here move fast and it is updated as they do.
