Nemotron 3 Ultra
The spec sheet
Figures as the maker publishes them.
Price
- Input, per 1M tokens
- $0.50
- Output, per 1M tokens
- $2.20
Memory
- Context window
- 262K tokens
- Longest answer
- 16K tokens
Abilities
- Reasoning
- Yes
- Reads images
- No
- Calls tools
- Yes
- Takes in
- text
- Gives back
- text
The model
- Lab
- NVIDIA
- Model id
- nvidia/nemotron-3-ultra-550b-a55b
- Released
- 4 Jun 2026
- Open weights
- Yes (nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16)
