Noometry

Model comparison

GPT-5.4 vs Qwen3 32B

GPT-5.4 is the stronger model overall, scoring 59.5 to 38.8 on the Noometry Index. Qwen3 32B costs 4.6× less per token, which makes it the better buy when GPT-5.4's lead doesn't matter for your workload.

Last verified . 23 shared benchmarks.

GPT-5.4 OpenAI

59.5

Rank #14 Confirmed

Qwen3 32B Alibaba (Qwen)

38.8

Rank #166 Confirmed

Summary

  • They share 23 benchmarks with published results for both. GPT-5.4 scores higher in 9 categories and Qwen3 32B in 0 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where GPT-5.4 leads 61.7 to 19.6.
  • The biggest single-benchmark swing is Chess Puzzles: 44% for GPT-5.4 and 5% for Qwen3 32B.
  • Qwen3 32B is cheaper at $0.70 / $2.80 per million input/output tokens, against $2.50 / $15 for GPT-5.4.
  • GPT-5.4 accepts more context: 1.05M tokens versus 131K.
  • Qwen3 32B has downloadable open weights; the other is API-only.

Side by side

GPT-5.4Qwen3 32B
ProviderOpenAIAlibaba (Qwen)
Noometry Index59.538.8
Released2026-03-052025-04
WeightsProprietaryOpen
Context window1.05M131K
Max output128K16K
Input $ / M tokens$2.50$0.70
Output $ / M tokens$15$2.80
Results tracked6826

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

Category by category

Coding GPT-5.4 leads

GPT-5.4: 52.8 (#28), Qwen3 32B: 37.3 (#184)

BenchmarkGPT-5.4Qwen3 32B
SciCode56.6%35.4%
LMArena Coding14971358
SWE-bench Verified76.9%—
DeepSWE51.8%—
Aider Polyglot—40%
LMArena WebDev1465—
GSO31.4%—
WeirdML77.7%—
MirrorCode15.6%—
ALE-Bench1,607—
AlgoTune1.85—

Agentic & Tool Use GPT-5.4 leads

GPT-5.4: 46.5 (#12), Qwen3 32B: 32.3 (#58)

BenchmarkGPT-5.4Qwen3 32B
Terminal-Bench81.8%—
APEX-Agents52.4%—
Berkeley Function Calling Leaderboard—48.7%
τ²-bench Banking39.4%—
DeepResearch Bench35.1%—
PostTrainBench19%—
GBAEval45.1%—
LMArena Search1197—
METR Time Horizons74.3%—
Vending-Bench 26,144—

Reasoning GPT-5.4 leads

GPT-5.4: 61.7 (#17), Qwen3 32B: 19.6 (#233)

BenchmarkGPT-5.4Qwen3 32B
Kagi LLM Benchmark63.8%54.9%
CritPt23.4%0.3%
Chess Puzzles44%5%
LMArena Hard Prompts14851335
DTBench94.4%67.5%
LMCA52%17.3%
Epoch Capabilities Index156.92138.51
ARC-AGI-274%—
NYT Connections (extended)91.3%—
ARC-AGI-193.7%—
EnigmaEval16%—
Thematic Generalization80%—
EBR-Bench25.4%—
Mystery Game Puzzles37%—
ForecastBench59.5—

Math GPT-5.4 leads

GPT-5.4: 73.3 (#14), Qwen3 32B: 37.9 (#89)

BenchmarkGPT-5.4Qwen3 32B
OTIS Mock AIME 2024-202597.8%66.9%
LMArena Math14881399
FrontierMath (Tiers 1-3)78.6%—
FrontierMath Tier 449%—
MathArena Final-Answer Competitions83.1%—
ProofBench56%—
FrontierMath (Feb 2025 set)47.6%—
FrontierMath Tier 4 (v1)27.1%—

Knowledge GPT-5.4 leads

GPT-5.4: 65.3 (#13), Qwen3 32B: 40.0 (#112)

BenchmarkGPT-5.4Qwen3 32B
GPQA Diamond93.3%65.7%
Vectara Hallucination Rate7%5.9%
LMArena Expert15071362
Humanity's Last Exam36.2%—
SimpleQA Verified45.1%—

Multimodal Not comparable

GPT-5.4: 43.8 (#19), Qwen3 32B: —

BenchmarkGPT-5.4Qwen3 32B
LMArena Vision1302—
Blueprint-Bench 227.1%—
Furniture Assembly37.5%—
LMArena Document1471—

Multilingual GPT-5.4 leads

GPT-5.4: 56.4 (#22), Qwen3 32B: 45.8 (#162)

BenchmarkGPT-5.4Qwen3 32B
LMArena Non-English14651317
LMArena Chinese15201357
LMArena German14721341
LMArena Russian14801311
LMArena French1493—
LMArena Japanese1484—
LMArena Korean1449—
LMArena Spanish1453—

Instruction Following GPT-5.4 leads

GPT-5.4: 77.3 (#25), Qwen3 32B: 69.1 (#174)

BenchmarkGPT-5.4Qwen3 32B
LMArena Instruction Following14691305

Long Context GPT-5.4 leads

GPT-5.4: 50.3 (#8), Qwen3 32B: 43.7 (#83)

BenchmarkGPT-5.4Qwen3 32B
LMArena Longer Query14741327
Fiction.LiveBench—74.2%
CL-bench27.9%—
CL-bench Life21.7%—

Writing & Preference GPT-5.4 leads

GPT-5.4: 72.1 (#18), Qwen3 32B: 53.1 (#158)

BenchmarkGPT-5.4Qwen3 32B
LMArena Text14691340
LMArena Creative Writing14391297
LMArena Multi-Turn14821331
EQ-Bench Creative Writing1840—
EQ-Bench 41272—

Frequently asked questions

Is GPT-5.4 better than Qwen3 32B?

GPT-5.4 is the stronger model overall, scoring 59.5 to 38.8 on the Noometry Index. Qwen3 32B costs 4.6× less per token, which makes it the better buy when GPT-5.4's lead doesn't matter for your workload.

Which is cheaper, GPT-5.4 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 lists at $2.50 and $15.

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

GPT-5.4 scores higher on coding benchmarks: 52.8 versus 37.3 in the Noometry coding category.

Which has the bigger context window?

GPT-5.4 does, with 1.05M tokens against 131K.

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

23 benchmarks have published results for both models. GPT-5.4 has 68 scored results on Noometry and Qwen3 32B has 26.

Related comparisons

Go deeper