Skip to main content
Back to Explore

Qwen2.5 Instruct 72B

Alibaba·Released 2024-09-19
Open SourceMoE

Pricing

Input

$0.36

per 1M tokens

Output

$0.40

per 1M tokens

Blended

$0.37

per 1M tokens

Cheaper than 58% of models. Median price is $0.54/1M tokens.

Cost Calculator

Tokens per day1M
100K100M

Daily

$0.37

Monthly

$11.10

vs. Similar Models

GPT-4o (Aug '24)Q:0.0
$4.38+1082%
Qwen3 Omni 30B A3B (Reasoning)Q:0.0
$0.43+16%
Ling-flash-2.0Q:+0.1
$0.25-33%
Perplexity: SonarQ:-0.1
$1.00+170%

Performance

56

tokens/sec

Faster than 25% of models

1.23

seconds

Faster than 43% of models

1.23

seconds

Faster than 62% of models

Market Median

94 tok/s

41% slower

Median TTFT

1.10s

11% slower

Throughput/Dollar

150

tok/s per $/1M

Speed Comparison

Pixtral Large
56 tok/s+0%
Mistral Large 3
56 tok/s-0%
Qwen3.5 Omni Plus
55 tok/s-1%

Benchmarks

MMLU-Pro
72.0%
GPQA Diamond
49.1%
HLE
4.2%
LiveCodeBench
27.6%
SciCode
26.7%
TerminalBench Hard
4.5%
MATH-500
85.8%
AIME
16.0%
AIME 2025
14.0%
IFBench
36.9%
Long Context Recall
20.3%
Tau2
34.5%
Market AverageTop Score

Open Source

Quick Compare

Similar Models

Compare all 7 models