Skip to main content
Back to Explore

Llama Nemotron Super 49B v1.5 (Non-reasoning)

NVIDIA·Released 2025-07-25
Open Source

Pricing

Input

$0.10

per 1M tokens

Output

$0.40

per 1M tokens

Blended

$0.17

per 1M tokens

Cheaper than 72% of models. Median price is $0.54/1M tokens.

Cost Calculator

Tokens per day1M
100K100M

Daily

$0.17

Monthly

$5.25

vs. Similar Models

Llama 3.3 Instruct 70BQ:-0.1
$0.61+250%
Mistral Small 3.1Q:-0.1
$0.14-21%
OpenAI: GPT-4o (2024-05-13)Q:-0.1
$7.50+4186%
Qwen3 32B (Non-reasoning)Q:-0.1
$0.26+49%

Performance

52

tokens/sec

Faster than 21% of models

0.26

seconds

Faster than 98% of models

0.26

seconds

Faster than 100% of models

Market Median

94 tok/s

44% slower

Median TTFT

1.10s

77% faster

Throughput/Dollar

298

tok/s per $/1M

Speed Comparison

Llama Nemotron Super 49B v1.5 (Reasoning)
52 tok/s-0%
Qwen: Qwen3.5 397B A17B
52 tok/s-0%
MiniMax: MiniMax M2.7
52 tok/s+0%

Benchmarks

MMLU-Pro
69.2%
GPQA Diamond
48.1%
HLE
4.3%
LiveCodeBench
29.0%
SciCode
23.8%
TerminalBench Hard
3.8%
MATH-500
77.0%
AIME
13.7%
AIME 2025
8.0%
IFBench
32.9%
Long Context Recall
22.0%
Tau2
25.1%
Market AverageTop Score

Open Source

Quick Compare

Similar Models

Compare all 7 models