AI Rate CardNVIDIA Nemotron API pricing: Nemotron 3 Ultra, Super, Nano and 3.5 at every provider
Updated Nov 5, 2026
Nemotron is NVIDIA's family of open models, from small Nano models to the 550B Ultra. NVIDIA serves them on its own build platform, and inference clouds and API resellers sell them too - over thirty providers carry at least one Nemotron. Several give some Nemotron models away for free. Below: the typical price (median across providers) and live tables of every seller, rebuilt from today's prices on every page view.
The models in one list
Price per 1M tokens, input / output, in the 4 November 2026 collection; "typical" is the median across providers:
- Nemotron 3 Ultra 550B-A55B - typically $0.60 / $2.40, input from $0.37; 1M tokens of context, 16 providers. The largest, a mixture-of-experts model with 55B active parameters.
- Nemotron 3 Super 120B-A12B - typically $0.12 / $0.55, input from $0.07; 1M context, 15 providers.
- Nemotron 3 Nano 30B-A3B - typically $0.06 / $0.24, input from $0.04; 1M context, 14 providers. Small and cheap.
- Nemotron 3.5 Super VL - typically $0.15 / $0.50, 14 providers: takes images as well as text.
- Nemotron 3.5 Lightning - typically $0.06 / $0.20, input from $0.03; 12 providers, and free at 8 of them.
- Nemotron 3.5 Content Safety - $0.20 / $0.20, 8 providers: a moderation model, see moderation and safety models.
Nemotron 3 Ultra: every provider
| Provider | Model | Input, $ | Output, $ |
|---|---|---|---|
| nvidia | Nemotron 3 Ultra 550B A55B | Free | Free |
| aihubmix | Nemotron 3 Ultra 550B A55B (free) | Free | Free |
| kilo | NVIDIA: Nemotron 3 Ultra (free) | Free | Free |
| opencode | Nemotron 3 Ultra Free | Free | Free |
| openrouter | NVIDIA: Nemotron 3 Ultra (free) | Free | Free |
| ollama-cloud | nemotron-3-ultra | $0.10 | $3.00 |
| ollama | Nemotron 3 Ultra 550B A55B | $0.10 | $3.00 |
| requesty | Nemotron 3 Ultra 550B A55B | $0.37 | $1.87 |
| kilo | NVIDIA: Nemotron 3 Ultra | $0.40 | $1.76 |
| deepinfra | Nemotron 3 Ultra 550B A55B | $0.50 | $2.20 |
| edenai | Nemotron 3 Ultra 550B A55B | $0.50 | $2.20 |
| fastrouter | NVIDIA: Nemotron 3 Ultra 550B A55B | $0.50 | $2.20 |
| requesty | Nemotron 3 Ultra 550B A55B | $0.50 | $2.50 |
| trustedrouter | NVIDIA: Nemotron 3 Ultra | $0.53 | $2.27 |
| baseten | Nemotron 3 Ultra 550B A55B | $0.60 | $2.40 |
Live data: this table is built from today's prices.
Nemotron 3 Super: every provider
| Provider | Model | Input, $ | Output, $ |
|---|---|---|---|
| nvidia | Nemotron 3 Super | Free | Free |
| fastrouter | NVIDIA: Nemotron 3 Super 120B (Free) | Free | Free |
| opencode | Nemotron 3 Super Free | Free | Free |
| ollama-cloud | nemotron-3-super | $0.015 | $0.60 |
| ollama | Nemotron 3 Super 120B A12B | $0.015 | $0.60 |
| requesty | Nemotron 3 Super 120B A12B | $0.068 | $0.32 |
| deepinfra | Nemotron 3 Super 120B A12B | $0.075 | $0.40 |
| edenai | Nemotron 3 Super 120B A12B | $0.075 | $0.40 |
| fastrouter | NVIDIA: Nemotron 3 Super 120B A12B | $0.075 | $0.40 |
| kilo | NVIDIA: Nemotron 3 Super | $0.075 | $0.40 |
| openrouter | NVIDIA: Nemotron 3 Super | $0.075 | $0.40 |
| trustedrouter | NVIDIA: Nemotron 3 Super | $0.07913 | $0.42 |
| kenari | Nemotron 3 Super 120B A12B | $0.09318 | $0.46 |
| aiml | Nemotron 3 Super 120B A12B | $0.09522 | $0.48 |
| requesty | Nemotron 3 Super 120B A12B | $0.10 | $0.50 |
Live data: this table is built from today's prices.
How to choose
- The strongest Nemotron: 3 Ultra - still about a quarter of the price of the large closed models.
- Good value: 3 Super, a fifth of Ultra's price.
- Volume at the lowest price: 3 Nano or 3.5 Lightning - Lightning is also free at several providers, usually with rate limits.
- Images and documents: 3.5 Super VL.
- Other cheap open models: gpt-oss, Gemma, the cheapest LLM APIs.
- Every Nemotron offer in one list: Nemotron in the catalog.
All prices are in USD per 1 million tokens, without taxes, as the providers publish them; each row links to its card with the source and the date it was checked.