Skip to main content
Back to Explore

NVIDIA Nemotron 3 Super 120B A12B (Reasoning)

NVIDIA·Released 2026-03-11
Open Source

Pricing

Input

$0.30

per 1M tokens

Output

$0.75

per 1M tokens

Blended

$0.41

per 1M tokens

Cheaper than 55% of models. Median price is $0.54/1M tokens.

Cost Calculator

Tokens per day1M
100K100M

Daily

$0.41

Monthly

$12.36

vs. Similar Models

DeepSeek V3.2 Exp (Reasoning)Q:0.0
$0.31-25%
Inception: Mercury 2Q:-0.1
$0.38-9%
Claude 4 Opus (Non-reasoning)Q:+0.1
$30.00+7182%
Claude 4 Sonnet (Non-reasoning)Q:+0.1
$6.00+1356%

Performance

232

tokens/sec

Faster than 93% of models

1.04

seconds

Faster than 55% of models

9.67

seconds

Faster than 41% of models

Market Median

94 tok/s

146% faster

Median TTFT

1.11s

7% faster

Throughput/Dollar

563

tok/s per $/1M

Speed Comparison

OpenAI: gpt-oss-20b
232 tok/s+0%
LFM2.5-8B-A1B
231 tok/s-0%
xAI: Grok 4.20
234 tok/s+1%

Benchmarks

MMLU-ProNot evaluated
GPQA Diamond
80.0%
HLE
19.2%
LiveCodeBenchNot evaluated
SciCode
36.0%
TerminalBench Hard
28.8%
MATH-500Not evaluated
AIMENot evaluated
AIME 2025Not evaluated
IFBench
71.5%
Long Context Recall
60.0%
Tau2
67.8%
Market AverageTop Score

Open Source

Quick Compare

Similar Models

Compare all 7 models