Back to models
available
Nemotron 3 Ultra 550B A55B
nvidia/nemotron-3-ultra-nvfp4Prices
- Input
$0.60 per 1M tokens
high- Input (cached read, priority)
$0.15 per 1M tokens
low- Input (cached read)
$0.12 per 1M tokens
high- Input (priority)
$0.75 per 1M tokens
low- Output
$2.40 per 1M tokens
high- Output (priority)
$3.00 per 1M tokens
low
Specifications
- Provider
- nvidia
- Context window
- 262,144
- Max output tokens
- 32,768
- Open weights
- yes
- Released
- 06/04/2026
- Retires
- no date announced
- First seen by this tracker
Price history
No price of this model has changed since we started tracking it.
- Input
- 0.6 per 1M tokens, unchanged since 09/14/2026
- Input (cached read)
- 0.12 per 1M tokens, unchanged since 09/14/2026
- Output
- 2.4 per 1M tokens, unchanged since 09/14/2026
- Input (cached read, priority)
- 0.15 per 1M tokens, unchanged since 10/01/2026
- Input (priority)
- 0.75 per 1M tokens, unchanged since 10/01/2026
- Output (priority)
- 3 per 1M tokens, unchanged since 10/01/2026