Skip to main content
Back to Explore

Llama 3.1 Nemotron Instruct 70B

NVIDIA·Released 2024-10-15
Open Source

Pricing

Input

$1.20

per 1M tokens

Output

$1.20

per 1M tokens

Blended

$1.20

per 1M tokens

Cheaper than 33% of models. Median price is $0.54/1M tokens.

Cost Calculator

Tokens per day1M
100K100M

Daily

$1.20

Monthly

$36.00

vs. Similar Models

Cohere: Command AQ:+0.1
$4.38+265%
Nova ProQ:+0.1
$1.40+17%
NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)Q:-0.2
$0.09-93%
NVIDIA Nemotron Nano 9B V2 (Non-reasoning)Q:-0.2
$0.09-93%

Performance

276

tokens/sec

Faster than 96% of models

4.14

seconds

Faster than 17% of models

4.14

seconds

Faster than 47% of models

Market Median

94 tok/s

193% faster

Median TTFT

1.11s

273% slower

Throughput/Dollar

230

tok/s per $/1M

Speed Comparison

NVIDIA Nemotron Nano 12B v2 VL (Reasoning)
273 tok/s-1%
gpt-oss-20B (low)
269 tok/s-3%
Nova Micro
262 tok/s-5%

Benchmarks

MMLU-Pro
69.0%
GPQA Diamond
46.5%
HLE
4.6%
LiveCodeBench
16.9%
SciCode
23.3%
TerminalBench Hard
4.5%
MATH-500
73.3%
AIME
24.7%
AIME 2025
11.0%
IFBench
30.7%
Long Context Recall
7.0%
Tau2
23.1%
Market AverageTop Score

Open Source

Quick Compare

Similar Models

Compare all 7 models