NVIDIA · Model
Nemotron 3 Super pricing, context window and specs
Nemotron 3 Super costs $0.08 per million input tokens and $0.45 per million output tokens, and reads up to 262,144 tokens per request. By blended price it is number 3 of NVIDIA’s 4 current models here, counting from the cheapest.
- Input / 1M
- $0.08
- Output / 1M
- $0.45
- Cached / 1M
- –
- Context
- 262,144
- Max output
- Not published
- Released
- 2026-03-11
Estimate your cost with Nemotron 3 SuperCompare all NVIDIA modelsCount tokens for Nemotron 3 Super
Pricing
Nemotron 3 Super API pricing
| Price per 1M tokens | Input | Output |
|---|---|---|
| Standard | $0.08 | $0.45 |
Output costs 5.6× the input price, so long replies drive the bill more than long prompts. No cached-input price is published, so repeated prompt prefixes are billed at the full input price.
History
Nemotron 3 Super price history
Price increase, +5.3% blended
- Input
- from $0.085 to $0.08
- −5.9%
- Output
- from $0.40 to $0.45
- +12%
Nemotron 3 Super has open weights, so this is the price of OpenRouter’s top-ranked host, which can move when hosts change rather than because NVIDIA changed its price. Every change across all models is in the AI pricing changes log.
Examples
What Nemotron 3 Super costs in practice
| Workload | Tokens in / out | Requests/day | Per request | Per month |
|---|---|---|---|---|
| Support chatbot | 1,500 / 400 | 1,000 | $0.0003 | $9.13 |
| RAG app | 6,000 / 500 | 500 | $0.000705 | $10.72 |
| Coding agent | 40,000 / 2,000 | 200 | $0.0041 | $24.94 |
| Document summariser | 8,000 / 600 | 300 | $0.00091 | $8.30 |
Change any number in the cost calculator.
Context
Context window
Nemotron 3 Super accepts up to 262,144 tokens per request, roughly 196,608 English words or 393 pages. That budget is shared between your prompt and the reply, and no separate cap on the reply is published.
Filling the whole window with one prompt costs about $0.021 at list price, so for repeated long-document work, retrieval (sending only the relevant parts) or prompt caching is usually much cheaper.
Features
Capabilities
- Image input
- No
- Tool / function calling
- Yes
- Reasoning mode
- Yes
- Prompt caching
- No
- Structured output
- Yes
- Open weights
- Yes
- Accepts
- text
- Knowledge cutoff
- Not published
As declared by the provider’s API. It shows a feature exists, not how well it works.
API
Nemotron 3 Super model ID
- OpenRouter
nvidia/nemotron-3-super-120b-a12b- Open weights
- nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8
Use the OpenRouter ID when calling it through OpenRouter; NVIDIA’s own API may name it differently. See the model ID reference for the providers we document.
NVIDIA
Where Nemotron 3 Super sits in NVIDIA’s line-up
Ranked by blended price (three parts input to one part output, $0.1725 per 1M for Nemotron 3 Super), it is number 3 of NVIDIA’s 4 current models here, counting from the cheapest. One step down, Nemotron 3 Nano 30B A3B costs $0.105 blended (39% less). It has the same published limits and features as Nemotron 3 Super. One step up, Nemotron 3 Ultra costs $1.05 (6.1× as much). Unlike Nemotron 3 Super, it reads up to 202,800 tokens (262,144 here). See every NVIDIA model and price
| Model | Input / 1M | Output / 1M | Context | Released |
|---|---|---|---|---|
| Nemotron 3.5 Lightning | $0.07 | $0.20 | 262,144 | 2026-08-11 |
| Nemotron 3 Nano 30B A3B | $0.06 | $0.24 | 262,144 | 2025-12-14 |
| Nemotron 3 Super (this page) | $0.08 | $0.45 | 262,144 | 2026-03-11 |
| Nemotron 3 Ultra | $0.60 | $2.40 | 202,800 | 2026-06-04 |
NVIDIA’s current models with a page here, cheapest first. Older and superseded models are listed on the provider page and in the comparison table.
Save
Cheaper alternatives
- Nemotron 3.5 LightningNVIDIA41% cheaper
- Gemma 4 26B A4BGoogle38% cheaper
- Qwen3.5-9BQwen35% cheaper
- Seed 1.6 FlashByteDance Seed24% cheaper
Current models from major providers, one per provider, that keep tool calling, a reasoning mode, a usable context window and output limit. Compared by blended price.
Compare
Similarly priced models
- Seed-2.0-MiniByteDance Seed$0.10 / $0.40
- Gemma 4 31BGoogle$0.09 / $0.34
- Ministral 3 8B 2512Mistral$0.15 / $0.15
- GPT-6 LunaOpenAI$0.10 / $0.50
- Claude Haiku 5.5Anthropic$0.10 / $0.50
Current models from other major providers, closest blended price.
FAQ
Nemotron 3 Super questions
How much does Nemotron 3 Super cost?
Nemotron 3 Super costs $0.08 per million input tokens and $0.45 per million output tokens. A typical chatbot reply (1,500 tokens in, 400 out) costs about $0.0003. Use the LLM cost calculator for your own workload.
What is Nemotron 3 Super’s context window?
262,144 tokens, about 196,608 English words or 393 printed pages, shared between your prompt and the reply. No separate limit on the reply is published, so the window is the limit.
Does Nemotron 3 Super support prompt caching and batch requests?
No cached-input price is published for it. No batch price is published for it.
What are cheaper alternatives to Nemotron 3 Super?
With the same essentials (tool calling), the cheapest options are Nemotron 3.5 Lightning (41% cheaper), Gemma 4 26B A4B (38% cheaper), Qwen3.5-9B (35% cheaper). Cheaper doesn’t mean equivalent, so test them on your own prompts.
What is the API model ID for Nemotron 3 Super?
On OpenRouter it is nvidia/nemotron-3-super-120b-a12b. NVIDIA’s own API may use a different name, so check its model list before you deploy. Pass the ID as the model value in each request.
Is Nemotron 3 Super open source?
Yes, Nemotron 3 Super has open weights, so you can run it yourself or choose between several hosting providers. The prices here are a typical hosted price.