Pricing · per 1M tokens
| Cached input | $0.120/Mtok (20% of input — prompt caching) |
|---|---|
| Blended avg cost* | $2.10/Mtok |
Available on 3 hosts · cheapest first
NVIDIA Nemotron 3 Ultra is sold by 3 providers. Prices are per 1M tokens (blended = avg of input & output). The first-party row is the model maker; “vs first-party” shows each host’s blended price relative to it.
| Host | Input | Output | Blended | vs first-party |
|---|---|---|---|---|
| Fireworks details → | $0.600 | $2.40 | $1.50 | -29% |
| NVIDIA first-party | $0.600 | $3.60 | $2.10 | — |
| Together details → | $0.600 | $3.60 | $2.10 | 0% |
Each host links to its provider page and official pricing. Prices refresh twice daily.
Overview
NVIDIA's Nemotron 3 Ultra reasoning model. $0.60/$3.60 per 1M on hosted platforms.
Deploy this open model on rented GPUs
Open weights mean you can self-host instead of paying per-token API prices. These platforms let you serve it on demand — often cheaper at scale.
Partner links — we may earn a commission if you sign up, at no cost to you. We only list platforms we'd recommend regardless.
Capabilities
Benchmark performance
- LMSYS Chatbot Arena: 1420.5 ELO (human preference)
- Artificial Analysis Intelligence Index: 37.8
Source: Artificial Analysis · Updated Jun 29, 2026 · See full rankings →
Specifications
| Provider | NVIDIA |
|---|---|
| Context window | 128K tokens |
| Modality | text |
| Parameters | Proprietary |
| Open source | Yes — open weights available |
| Released | Jan 1, 2026 |
| Status | Current |
| Last updated | Jun 25, 2026 |
| Tags | open-weights mid-tier reasoning |
Cost calculator
At a glance
Embed this price
A live, auto-updating badge for your README or docs:
Price history
Last updated: Jul 10, 2026 · price links to the official provider pricing page (source above)
No price changes detected — this model's pricing has been stable since we began tracking it.