Nemotron 3 Ultra
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...
Leaderboards
Average Score combining domain-specific Autobench scores; Higher is better
Performance vs. Industry Average
Intelligence
Nemotron 3 Ultra is of lower intelligence compared to average (4.5), with an intelligence score of 4.5.
Price
Nemotron 3 Ultra is cheaper compared to average ($1.13 per 1M Tokens) with a price of $0.69 per 1M Tokens.
Latency
Nemotron 3 Ultra has a lower average latency compared to average (47.46s), with an average latency of 38.07s.
P99 Latency
Nemotron 3 Ultra has a higher P99 latency compared to average (288.94s), taking 322.09s to receive the first token at P99 (TTFT).
Context Window
Nemotron 3 Ultra has a larger context window than average (453k tokens), with a context window of 512k tokens.