NVIDIA cloud models

19 models tracked via Artificial Analysis. Compare cloud performance, then find local GGUF versions in the GraySoft model catalog.

Model	Intelligence	Speed (tok/s)
Nemotron 3 Ultra 550B A55B (Reasoning)	29.6	156.367
Nemotron 3 Super 120B A12B (Reasoning)	18.6	141.927
Nemotron 3.5 Lightning	16.4	295.608
Nemotron Cascade 2 30B A3B	11.5	0
Nemotron 3 Nano Omni 30B A3B Reasoning	9	321.434
NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)	8.6	142.767
Llama Nemotron Super 49B v1.5 (Reasoning)	6.6	18.646
Llama 3.3 Nemotron Super 49B v1 (Reasoning)	6.4	0
Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)	3.4	52.821
NVIDIA Nemotron Nano 12B v2 VL (Reasoning)	3.3	25.47
NVIDIA Nemotron Nano 9B V2 (Reasoning)	3.2	94.506
NVIDIA Nemotron 3 Nano 4B	3.1	0
Llama Nemotron Super 49B v1.5 (Non-reasoning)	3	23.094
Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)	2.9	0
Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)	2.8	0
Llama 3.1 Nemotron Instruct 70B	2.1	19.298
NVIDIA Nemotron Nano 9B V2 (Non-reasoning)	1.8	136.939
NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)	1.8	165.785
NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)	1	72.097

Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.