Nemotron 3.5 Super VL on OpenRouter is the easiest way to plug Nvidia’s new vision model into almost any app or agent: one API, a free variant, and a cheap paid version with a 1M-token context. My video below shows it in action. Here are the model IDs, the pricing, and how to use it in Hermes and other tools.
Key takeaways
- Paid model id:
nvidia/nemotron-3.5-super-vl. - Free model id:
nvidia/nemotron-3.5-super-vl:free. - Paid pricing: $0.15/M input, $0.50/M output tokens.
- Context: about 1M tokens.
Nemotron 3.5 Super VL on OpenRouter
| Version | Model id | Context | Price |
|---|---|---|---|
| Paid | nvidia/nemotron-3.5-super-vl | 1,048,576 tokens | $0.15/M input, $0.50/M output |
| Free | nvidia/nemotron-3.5-super-vl:free | ~1,000,000 tokens | Free (rate limited) |
Prices are from OpenRouter’s model list at the time of writing and can change.
How to use it with OpenRouter
- Create an OpenRouter account and API key.
- In your app or agent, choose OpenRouter as the provider.
- Enter the model id — free or paid.
- Send text, images or document pages.
Any agent that lets you pick a model from OpenRouter can use it — you don’t need my exact setup.
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 →
Use it in Hermes Agent via OpenRouter
Run hermes model, choose OpenRouter, and select nvidia/nemotron-3.5-super-vl:free. I keep it in its own profile so I can switch to it for visual jobs in one click. Full steps are in my Nemotron 3.5 Super VL Hermes Agent guide.
Free or paid?
Start with the free variant for testing and light use. Switch to paid when you hit rate limits or run big jobs — at $0.15 per million input tokens it’s one of the cheaper ways to get a capable vision model. Other Nemotron 3.5 models on OpenRouter include Lightning and a content-safety model, both with free variants too.
What it’s good at
- Reading screenshots, charts and document pages.
- Agentic tool use — 63.44% on Hermes Bench (Nvidia-reported).
- Long documents — strong long-context results.
- Fast responses thanks to its 12B-active design.
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.
Calling it from your own code
OpenRouter uses an OpenAI-compatible API, so most SDKs and tools work with a base-URL change. Point your client at OpenRouter’s API, set your OpenRouter key, and use nvidia/nemotron-3.5-super-vl as the model. To send an image, include it in the message content alongside your text, the same way you would with other vision models. Tool calling works too, which is what makes it useful inside agents.
Keeping costs predictable
Set a spending limit on your OpenRouter account, start with the free variant, and track usage for a week before scaling up. Because only 12B parameters are active per token, the paid version is cheap, but long contexts and images still add tokens. Cropping images and trimming conversation history keeps costs down without hurting results.
Other apps that work with it
Because it’s on OpenRouter, Nemotron 3.5 Super VL works in any app or agent that lets you pick an OpenRouter model: Hermes Agent, OpenClaw, OpenCode and many chat and automation tools. You don’t need my exact setup — if your tool supports OpenRouter, you can try it today.
Final thoughts
One more benefit of using it through OpenRouter: if Nvidia or another host releases an update, or if the free variant changes, you only need to change the model id in one place. OpenRouter also shows which providers are serving the model and their speeds, which helps if you need the fastest responses for an agent that’s waiting on each step.
And because the same key works for hundreds of other models, you can compare Nemotron with alternatives side by side without signing up anywhere else.
In short: grab a key, start free, and upgrade only when you need to. It’s one of the simplest ways to give any agent the ability to see.
How I’d test it this week
Create a key, add the free model to one agent, and run your most common visual task through it ten times. Check the answers, the speed and whether you hit any limits. That small test tells you whether to stay on free or move to paid.
If you use several tools, add it to all of them with the same key — one account, one bill, one model everywhere.
Free variant limits and when to upgrade
The free variant is perfect for testing and daily light use, but free models on OpenRouter are rate limited, and heavy agent workflows can hit those limits quickly — especially with images and long contexts. If your scheduled jobs start failing with rate-limit errors, switch them to the paid model id. At $0.15 per million input tokens and $0.50 per million output tokens, most small businesses can run real agent workloads on it for very little.
The bottom line on Nemotron 3.5 Super VL OpenRouter
Nemotron 3.5 Super VL on OpenRouter gives you a free and a cheap paid version of a strong vision model through one API. Grab a key and plug it into your agent today.
FAQ: nemotron 3.5 super vl openrouter
What is the OpenRouter model id for Nemotron 3.5 Super VL?
nvidia/nemotron-3.5-super-vl, and nvidia/nemotron-3.5-super-vl:free for the free version.
How much does it cost on OpenRouter?
$0.15 per million input tokens and $0.50 per million output tokens for the paid version.
Is there a free version on OpenRouter?
Yes — the :free variant, with rate limits.
What is the context length?
About 1M tokens.
Can I use it in Hermes through OpenRouter?
Yes — choose OpenRouter in hermes model and select the model id.
Which other Nemotron 3.5 models are on OpenRouter?
Nemotron 3.5 Lightning and Nemotron 3.5 Content Safety are also listed, each with a free variant at the time of writing.
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 openrouter — the facts here move fast and it is updated as they do.
