Noometry

Model comparison

GPT-4o mini vs Qwen1.5-32B

Qwen1.5-32B is the stronger model overall, scoring 30.0 to 25.4 on the Noometry Index.

Last verified . 21 shared benchmarks.

GPT-4o mini OpenAI

25.4

Rank #337 Confirmed

Qwen1.5-32B Alibaba (Qwen)

30.0

Rank #288 Confirmed

Summary

  • They share 21 benchmarks with published results for both. GPT-4o mini scores higher in 5 categories and Qwen1.5-32B in 3 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where Qwen1.5-32B leads 30.5 to 9.9.
  • The biggest single-benchmark swing is BigCodeBench Complete: 57.4% for GPT-4o mini and 42% for Qwen1.5-32B.
  • Qwen1.5-32B has downloadable open weights; the other is API-only.

Side by side

GPT-4o miniQwen1.5-32B
ProviderOpenAIAlibaba (Qwen)
Noometry Index25.430.0
Released2024-07-182024-02-04
WeightsProprietaryOpen
Context window128K—
Max output16K—
Input $ / M tokens$0.15—
Output $ / M tokens$0.60—
Results tracked6021

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding Qwen1.5-32B leads

GPT-4o mini: 21.6 (#330), Qwen1.5-32B: 31.1 (#278)

BenchmarkGPT-4o miniQwen1.5-32B
BigCodeBench Instruct46.1%32.3%
LMArena Coding12901155
BigCodeBench Complete57.4%42%
Aider Polyglot3.6%—
WeirdML11.8%—
LiveBench Coding43.1%—
HumanEval+83.5%—
MBPP+72.2%—

Agentic & Tool Use Not comparable

GPT-4o mini: 27.2 (#95), Qwen1.5-32B: —

BenchmarkGPT-4o miniQwen1.5-32B
BALROG17.4%—

Reasoning Qwen1.5-32B leads

GPT-4o mini: 8.7 (#343), Qwen1.5-32B: 20.9 (#210)

BenchmarkGPT-4o miniQwen1.5-32B
LMArena Hard Prompts12671131
ARC-AGI-20%—
SimpleBench10.7%—
Kagi LLM Benchmark28.8%—
Chess Puzzles0%—
LiveBench Reasoning32.8%—
Mystery Game Puzzles12%—
DTBench54.4%—
LiveBench Data Analysis50%—
LMCA10.4%—
Epoch Capabilities Index126.56—
LiveBench41.3%—
PIQA88.7%—

Math Qwen1.5-32B leads

GPT-4o mini: 9.9 (#307), Qwen1.5-32B: 30.5 (#207)

BenchmarkGPT-4o miniQwen1.5-32B
LMArena Math12671155
FrontierMath (Tiers 1-3)0.7%—
OTIS Mock AIME 2024-20256.9%—
Omni-MATH28%—
LiveBench Math36.3%—
MATH Level 552.6%—
GSM8K91.3%—

Knowledge GPT-4o mini leads

GPT-4o mini: 17.7 (#279), Qwen1.5-32B: 13.5 (#291)

BenchmarkGPT-4o miniQwen1.5-32B
GPQA Diamond37.7%30.7%
LMArena Expert12351126
MMLU81.8%74.4%
SimpleQA Verified8.3%—
MMLU-Pro60.3%—
Confabulations37.2%—
GPQA (HELM)36.8%—
BoolQ88.7%—

Multimodal Not comparable

GPT-4o mini: 25.3 (#117), Qwen1.5-32B: —

BenchmarkGPT-4o miniQwen1.5-32B
LMArena Vision1066—
Video-MME64.8%—
GeoBench64%—
VPCT34%—

Multilingual GPT-4o mini leads

GPT-4o mini: 42.2 (#194), Qwen1.5-32B: 31.6 (#254)

BenchmarkGPT-4o miniQwen1.5-32B
LMArena Non-English12661106
LMArena Chinese12641176
LMArena French12961100
LMArena German12721058
LMArena Japanese12161027
LMArena Korean11951007
LMArena Russian12751073
LMArena Spanish12761089

Instruction Following GPT-4o mini leads

GPT-4o mini: 62.1 (#234), Qwen1.5-32B: 58.0 (#260)

BenchmarkGPT-4o miniQwen1.5-32B
LMArena Instruction Following12591116
LiveBench Instruction Following56.8%—
IFEval78.2%—

Long Context GPT-4o mini leads

GPT-4o mini: 39.1 (#181), Qwen1.5-32B: 34.7 (#241)

BenchmarkGPT-4o miniQwen1.5-32B
LMArena Longer Query12891146

Writing & Preference GPT-4o mini leads

GPT-4o mini: 39.7 (#243), Qwen1.5-32B: 34.4 (#266)

BenchmarkGPT-4o miniQwen1.5-32B
LMArena Text12861137
LMArena Creative Writing12681083
LMArena Multi-Turn12851140
Short-Story Creative Writing67.2%—
EQ-Bench Creative Writing873—
WildBench79.1%—
LiveBench Language28.6%—

Frequently asked questions

Is GPT-4o mini better than Qwen1.5-32B?

Qwen1.5-32B is the stronger model overall, scoring 30.0 to 25.4 on the Noometry Index.

Is GPT-4o mini or Qwen1.5-32B better for coding?

Qwen1.5-32B scores higher on coding benchmarks: 31.1 versus 21.6 in the Noometry coding category.

How many benchmarks do GPT-4o mini and Qwen1.5-32B share?

21 benchmarks have published results for both models. GPT-4o mini has 60 scored results on Noometry and Qwen1.5-32B has 21.

Related comparisons

Go deeper