Skip to main content
Back to Explore

NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)

NVIDIA·Released 2025-12-15
Open Source

Pricing

Input

$0.06

per 1M tokens

Output

$0.22

per 1M tokens

Blended

$0.10

per 1M tokens

Cheaper than 84% of models. Median price is $0.54/1M tokens.

Cost Calculator

Tokens per day1M
100K100M

Daily

$0.10

Monthly

$2.88

vs. Similar Models

Google: Gemini 2.5 FlashQ:-0.1
$0.85+785%
Upstage: Solar Pro 3Q:-0.1
$0.26+173%
Qwen: Qwen3 VL 235B A22B InstructQ:+0.1
$0.37+285%
GPT-5 mini (minimal)Q:+0.1
$0.69+617%

Performance

68

tokens/sec

Faster than 35% of models

2.65

seconds

Faster than 21% of models

31.98

seconds

Faster than 15% of models

Market Median

94 tok/s

28% slower

Median TTFT

1.11s

139% slower

Throughput/Dollar

711

tok/s per $/1M

Speed Comparison

Qwen: Qwen3.5-9B
68 tok/s-1%
Apriel-v1.6-15B-Thinker
69 tok/s+1%
DeepSeek: R1 Distill Llama 70B
67 tok/s-2%

Benchmarks

MMLU-Pro
79.4%
GPQA Diamond
75.7%
HLE
10.2%
LiveCodeBench
74.1%
SciCode
29.6%
TerminalBench Hard
13.6%
MATH-500Not evaluated
AIMENot evaluated
AIME 2025
91.0%
IFBench
71.1%
Long Context Recall
33.7%
Tau2
40.9%
Market AverageTop Score

Open Source

Quick Compare

Similar Models

Compare all 7 models