Nemotron 3 Ultra (550B A55B) by NVIDIA scores 58.1 out of 100 on the Neura Intelligence Index, ranking #152 among tracked AI models. This composite score is computed from 5 individual benchmarks across coding, math, reasoning, and language categories.

Nemotron 3 Ultra (550B A55B) costs $0.5/1M input tokens and $2.2/1M output tokens via API. At a typical 3:1 input-to-output ratio with 1M tokens/day usage, the estimated monthly cost is approximately $28.

Yes, Nemotron 3 Ultra (550B A55B) has open weights available under the openmdw license. You can download and self-host the model.

Nemotron 3 Ultra (550B A55B) supports a context window of 262K tokens, allowing it to process very long documents, codebases, and conversations in a single request.

Nemotron 3 Ultra (550B A55B) achieves its highest score on GPQA Diamond with a score of 87.0%. This benchmark evaluates 448 graduate-level science questions (biology, chemistry, physics) designed to be hard for non-experts.