Noometry

Model comparison

GPT-5.4 mini vs Qwen3 32B

GPT-5.4 mini is the stronger model overall, scoring 45.0 to 39.2 on the Noometry Index.

Last verified . 23 shared benchmarks.

GPT-5.4 mini OpenAI

45.0

Rank #76 Confirmed

Qwen3 32B Alibaba (Qwen)

39.2

Rank #172 Confirmed

Summary

  • They share 23 benchmarks with published results for both. GPT-5.4 mini scores higher in 7 categories and Qwen3 32B in 2 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where GPT-5.4 mini leads 51.5 to 40.0.
  • The biggest single-benchmark swing is LMCA: 40.8% for GPT-5.4 mini and 17.3% for Qwen3 32B.
  • Qwen3 32B is cheaper at $0.70 / $2.80 per million input/output tokens, against $0.75 / $4.50 for GPT-5.4 mini.
  • GPT-5.4 mini accepts more context: 400K tokens versus 131K.
  • Qwen3 32B has downloadable open weights; the other is API-only.

Side by side

GPT-5.4 miniQwen3 32B
ProviderOpenAIAlibaba (Qwen)
Noometry Index45.039.2
Released2026-03-172025-04
WeightsProprietaryOpen
Context window400K131K
Max output128K16K
Input $ / M tokens$0.75$0.70
Output $ / M tokens$4.50$2.80
Results tracked4626

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

Category by category

Coding GPT-5.4 mini leads

GPT-5.4 mini: 45.2 (#72), Qwen3 32B: 37.7 (#190)

BenchmarkGPT-5.4 miniQwen3 32B
SciCode49.9%35.4%
LMArena Coding14381358
FrontierCode27%—
Aider Polyglot—40%
LMArena WebDev1397—
WeirdML60.3%—
ALE-Bench1,189—

Agentic & Tool Use Qwen3 32B leads

GPT-5.4 mini: 29.9 (#81), Qwen3 32B: 32.6 (#62)

BenchmarkGPT-5.4 miniQwen3 32B
Berkeley Function Calling Leaderboard—48.7%
DeepResearch Bench36.3%—

Reasoning GPT-5.4 mini leads

GPT-5.4 mini: 30.4 (#85), Qwen3 32B: 20.2 (#241)

BenchmarkGPT-5.4 miniQwen3 32B
Kagi LLM Benchmark37.9%54.9%
CritPt10%0.3%
Chess Puzzles24%5%
LMArena Hard Prompts14241334
DTBench80%67.5%
LMCA40.8%17.3%
Epoch Capabilities Index148.84138.51
ARC-AGI-218.9%—
NYT Connections (extended)61.8%—
ARC-AGI-163.7%—
Thematic Generalization61.7%—
Mystery Game Puzzles11%—
ForecastBench57—

Math GPT-5.4 mini leads

GPT-5.4 mini: 45.5 (#75), Qwen3 32B: 39.7 (#99)

BenchmarkGPT-5.4 miniQwen3 32B
OTIS Mock AIME 2024-202588.9%66.9%
LMArena Math14191399
FrontierMath (Tiers 1-3)51.2%—
FrontierMath Tier 49.8%—
ProofBench21%—
FrontierMath (Feb 2025 set)28.3%—
FrontierMath Tier 4 (v1)2.1%—

Knowledge GPT-5.4 mini leads

GPT-5.4 mini: 51.5 (#67), Qwen3 32B: 40.0 (#125)

BenchmarkGPT-5.4 miniQwen3 32B
GPQA Diamond86.9%65.7%
Vectara Hallucination Rate5.5%5.9%
LMArena Expert14351362
SimpleQA Verified29.4%—

Multimodal Not comparable

GPT-5.4 mini: 39.7 (#56), Qwen3 32B: —

BenchmarkGPT-5.4 miniQwen3 32B
LMArena Vision1245—

Multilingual GPT-5.4 mini leads

GPT-5.4 mini: 51.9 (#96), Qwen3 32B: 45.6 (#167)

BenchmarkGPT-5.4 miniQwen3 32B
LMArena Non-English14051317
LMArena Chinese14461357
LMArena German14091341
LMArena Russian14171311
LMArena French1440—
LMArena Japanese1374—
LMArena Korean1368—
LMArena Spanish1405—

Instruction Following GPT-5.4 mini leads

GPT-5.4 mini: 74.1 (#102), Qwen3 32B: 68.9 (#179)

BenchmarkGPT-5.4 miniQwen3 32B
LMArena Instruction Following14051305

Long Context Too close to call

GPT-5.4 mini: 43.0 (#112), Qwen3 32B: 43.8 (#87)

BenchmarkGPT-5.4 miniQwen3 32B
LMArena Longer Query14071327
Fiction.LiveBench—74.2%

Writing & Preference GPT-5.4 mini leads

GPT-5.4 mini: 64.0 (#58), Qwen3 32B: 52.9 (#163)

BenchmarkGPT-5.4 miniQwen3 32B
LMArena Text14121340
LMArena Creative Writing13701297
LMArena Multi-Turn14291331
EQ-Bench Creative Writing1665—

Frequently asked questions

Is GPT-5.4 mini better than Qwen3 32B?

GPT-5.4 mini is the stronger model overall, scoring 45.0 to 39.2 on the Noometry Index.

Which is cheaper, GPT-5.4 mini or Qwen3 32B?

Qwen3 32B is cheaper. It lists at $0.70 per million input tokens and $2.80 per million output tokens; GPT-5.4 mini lists at $0.75 and $4.50.

Is GPT-5.4 mini or Qwen3 32B better for coding?

GPT-5.4 mini scores higher on coding benchmarks: 45.2 versus 37.7 in the Noometry coding category.

Which has the bigger context window?

GPT-5.4 mini does, with 400K tokens against 131K.

How many benchmarks do GPT-5.4 mini and Qwen3 32B share?

23 benchmarks have published results for both models. GPT-5.4 mini has 46 scored results on Noometry and Qwen3 32B has 26.

Related comparisons

Go deeper