NVIDIA
Nemotron 3 Ultra
nvidia/nemotron-3-ultra-550b-a55b
Input
$0.60
per 1M tokens
Output
$2.40
per 1M tokens
Cached input
$0.12
per 1M tokens
A free, rate-limited variant is also available.
NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE). Built on a hybrid Transformer-Mamba mixture-of-experts architecture, it...
- Released
- Context window
- 262K tokens
- Max output
- 183K tokens
- Input
- text
- Output
- text
- Reasoning
- Yes
Price history
No price changes since we started tracking on .
Providers on OpenRouter
5 endpoints serve this model. Uptime is OpenRouter's share of successful requests; prices per 1M tokens.
| Provider | Input | Output | Context window | Uptime 30 min | Uptime 24 h |
|---|---|---|---|---|---|
| DeepInfrafp4 · flexdiscount/priority tier | $0.40 | $1.76 | 262K | — | 99.41% |
| DeepInfrafp4 | $0.50 | $2.20 | 262K | 100% | 99.91% |
| BaseTenfp4 | $0.60 | $2.40 | 203K | 100% | 100% |
| BaseTenfp4 · us | $0.60 | $2.40 | 203K | 100% | 100% |
| Venicefp8 | $0.625 | $3.13 | 256K | 85.7% | 74.7% |