NVIDIA

Nemotron 3 Super

Type
language model
Context
1,000K tokens
Max output
262K
Released
11 March 2026
API string
nvidia/nemotron-3-super-120b-a12b
Weights
open

Prices by provider

Provider Input, $ per 1M Output, $ per 1M In our data since
NanoGPT $0.05 $0.25 31 Jul 2026
OpenRouter $0.085 $0.4 31 Jul 2026
Pioneer $0.09 $0.45 1 Aug 2026
Kilo Code $0.1 $0.5 31 Jul 2026
Vercel $0.15 $0.65 31 Jul 2026
NVIDIA $0.2 $0.8 31 Jul 2026
Weights & Biases $0.2 $0.8 1 Aug 2026
perplexity-agent $0.25 $2.5 31 Jul 2026
Cortecs $0.266 $0.799 31 Jul 2026
tensorx $0.3 $0.9 2 Aug 2026
Nebius Token Factory $0.3 $0.9 31 Jul 2026

Only the base tier and only per-token prices are shown. Batch, discounted and cached rates, as well as prices per image or per second of video, do not go into this table: they cannot stand in the same column as a price per million tokens.

The date in the last column is the day this price first entered our collection. The price may well be older: before that day we simply were not recording it. It has not changed since — otherwise a new row with a new date would stand in its place.

Measurement results

Task set Result Run conditions Measured by
Arena Score in Spanish 1,417.52 1,384–1,451
Arena Score in French 1,408.44 1,371–1,446
Arena Score, programming 1,405.97 1,392–1,420
Arena Score in English 1,400.01 1,390–1,410
Arena Score, expert questions 1,397.25 1,374–1,420
Arena Score, hard prompts 1,388.27 1,379–1,398
Arena Score, mathematics 1,382.77 1,358–1,408
Arena Score, overall 1,378.05 1,371–1,385
Arena Score, multi-turn dialogue 1,366.98 1,351–1,383
Arena Score, long queries 1,364.67 1,352–1,377
Arena Score, instruction following 1,349.75 1,337–1,363
Arena Score in Russian 1,336.89 1,316–1,357
Arena Score, creative writing 1,315.89 1,298–1,334