Qwen3 32B (Non-reasoning)

Alibaba·Released 2025-04-28

Open Source

Compare

Quality Index

14.5

274th of 440

Top 63%

Math Index

19.7

209th of 268

Top 78%

Price/1M

$1.23

503rd cheapest

295% above median

Top 75%

Speed

104 tok/s

Top 26%

TTFT

0.99s

Market Position

Qwen3 32B (Non-reasoning)Market Average

Pricing

Input

$0.70

per 1M tokens

Output

$2.80

per 1M tokens

Blended

$1.23

per 1M tokens

Cheaper than 25% of models. Median price is $0.31/1M tokens.

Cost Calculator

Tokens per day1M

100K100M

Daily

$1.23

Monthly

$36.75

vs. Similar Models

GPT-4o (May '24)Q:0.0

$7.50+512%

Gemini 2.0 Flash-Lite (Preview)Q:0.0

$0.00-100%

Llama 3.3 Instruct 70BQ:0.0

$0.64-48%

Mistral Small 3.1Q:0.0

$0.15-88%

Performance

104

tokens/sec

Faster than 74% of models

0.99

seconds

Faster than 30% of models

Market Median

46 tok/s

127% faster

Median TTFT

0.43s

130% slower

Throughput/Dollar

85

tok/s per $/1M

Speed Comparison

Qwen3 4B (Reasoning)

104 tok/s+0%

o1

104 tok/s+0%

Qwen3 32B (Reasoning)

105 tok/s+1%

Benchmarks

MMLU-Pro

72.7%

GPQA Diamond

53.5%

HLE

4.3%

LiveCodeBench

28.8%

SciCode

28.0%

TerminalBench HardNot evaluated

MATH-500

86.9%

AIME

30.3%

AIME 2025

19.7%

IFBench

31.5%

Long Context Recall

0.0%

Tau2Not evaluated

Market AverageTop Score

Open Source

Quick Compare

Similar Models

GPT-4o (May '24)

OpenAI

Q: 14.5$7.50/1M128K ctx

Slower: 34%Pricier: 512%

Llama 3.3 Instruct 70B