Noometry

Model comparison

GPT-5.4 vs Qwen3.7 Max

GPT-5.4 is the stronger model overall, scoring 59.4 to 51.5 on the Noometry Index. Qwen3.7 Max costs 1.5× less per token, which makes it the better buy when GPT-5.4's lead doesn't matter for your workload.

Last verified . 32 shared benchmarks.

GPT-5.4 OpenAI

59.4

Rank #16 Confirmed

Qwen3.7 Max Alibaba (Qwen)

51.5

Rank #42 Confirmed

Summary

  • They share 32 benchmarks with published results for both. GPT-5.4 scores higher in 8 categories and Qwen3.7 Max in 1 category; 7 gaps are clear of the uncertainty.
  • The widest gap is in agentic & tool use, where GPT-5.4 leads 46.5 to 22.1.
  • The biggest single-benchmark swing is GBAEval: 45.1% for GPT-5.4 and 0.4% for Qwen3.7 Max.
  • Qwen3.7 Max is cheaper at $2.50 / $7.50 per million input/output tokens, against $2.50 / $15 for GPT-5.4.
  • GPT-5.4 accepts more context: 1.05M tokens versus 1M.

Side by side

GPT-5.4Qwen3.7 Max
ProviderOpenAIAlibaba (Qwen)
Noometry Index59.451.5
Released2026-03-052026-05-19
WeightsProprietaryProprietary
Context window1.05M1M
Max output128K131K
Input $ / M tokens$2.50$2.50
Output $ / M tokens$15$7.50
Results tracked6833

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

Category by category

Coding GPT-5.4 leads

GPT-5.4: 52.6 (#33), Qwen3.7 Max: 50.4 (#45)

BenchmarkGPT-5.4Qwen3.7 Max
SWE-bench Verified76.9%77.3%
LMArena WebDev14651515
SciCode56.6%48.8%
LMArena Coding14971498
ALE-Bench1,6071,189
DeepSWE51.8%—
GSO31.4%—
WeirdML77.7%—
MirrorCode15.6%—
AlgoTune1.85—

Agentic & Tool Use GPT-5.4 leads

GPT-5.4: 46.5 (#13), Qwen3.7 Max: 22.1 (#135)

BenchmarkGPT-5.4Qwen3.7 Max
GBAEval45.1%0.4%
Terminal-Bench81.8%—
APEX-Agents52.4%—
τ²-bench Banking39.4%—
DeepResearch Bench35.1%—
PostTrainBench19%—
LMArena Search1197—
METR Time Horizons74.3%—
Vending-Bench 26,144—

Reasoning GPT-5.4 leads

GPT-5.4: 61.8 (#19), Qwen3.7 Max: 49.2 (#38)

BenchmarkGPT-5.4Qwen3.7 Max
NYT Connections (extended)91.3%85.1%
CritPt23.4%13.4%
Chess Puzzles44%19%
EBR-Bench25.4%9.5%
LMArena Hard Prompts14851483
Mystery Game Puzzles37%32%
DTBench94.4%92.3%
LMCA52%44%
Epoch Capabilities Index156.81153.68
ARC-AGI-274%—
SimpleBench—70.4%
Kagi LLM Benchmark63.8%—
ARC-AGI-193.7%—
EnigmaEval16%—
Thematic Generalization80%—
ForecastBench59.5—

Math GPT-5.4 leads

GPT-5.4: 73.5 (#19), Qwen3.7 Max: 62.4 (#32)

BenchmarkGPT-5.4Qwen3.7 Max
FrontierMath (Tiers 1-3)78.6%64.6%
FrontierMath Tier 449%34.1%
OTIS Mock AIME 2024-202597.8%95.6%
ProofBench56%26%
LMArena Math14881490
MathArena Final-Answer Competitions83.1%—
FrontierMath (Feb 2025 set)47.6%—
FrontierMath Tier 4 (v1)27.1%—

Knowledge GPT-5.4 leads

GPT-5.4: 65.3 (#14), Qwen3.7 Max: 61.6 (#28)

BenchmarkGPT-5.4Qwen3.7 Max
GPQA Diamond93.3%90.9%
SimpleQA Verified45.1%55.8%
LMArena Expert15071488
Humanity's Last Exam36.2%—
Vectara Hallucination Rate7%—

Multimodal Not comparable

GPT-5.4: 43.7 (#20), Qwen3.7 Max: —

BenchmarkGPT-5.4Qwen3.7 Max
LMArena Vision1303—
Blueprint-Bench 227.1%—
Furniture Assembly37.5%—
LMArena Document1471—

Multilingual Too close to call

GPT-5.4: 56.2 (#23), Qwen3.7 Max: 56.9 (#15)

BenchmarkGPT-5.4Qwen3.7 Max
LMArena Non-English14651474
LMArena Chinese15191530
LMArena Russian14801484
LMArena French1493—
LMArena German1472—
LMArena Japanese1485—
LMArena Korean1448—
LMArena Spanish1454—

Instruction Following Too close to call

GPT-5.4: 77.1 (#27), Qwen3.7 Max: 76.7 (#38)

BenchmarkGPT-5.4Qwen3.7 Max
LMArena Instruction Following14691460

Long Context GPT-5.4 leads

GPT-5.4: 50.3 (#8), Qwen3.7 Max: 45.4 (#40)

BenchmarkGPT-5.4Qwen3.7 Max
LMArena Longer Query14731482
CL-bench27.9%—
CL-bench Life21.7%—

Writing & Preference GPT-5.4 leads

GPT-5.4: 71.9 (#17), Qwen3.7 Max: 65.0 (#54)

BenchmarkGPT-5.4Qwen3.7 Max
LMArena Text14691476
LMArena Creative Writing14391449
EQ-Bench 412721110
LMArena Multi-Turn14821481
EQ-Bench Creative Writing1840—

Frequently asked questions

Is GPT-5.4 better than Qwen3.7 Max?

GPT-5.4 is the stronger model overall, scoring 59.4 to 51.5 on the Noometry Index. Qwen3.7 Max costs 1.5× 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 Qwen3.7 Max?

Qwen3.7 Max is cheaper. It lists at $2.50 per million input tokens and $7.50 per million output tokens; GPT-5.4 lists at $2.50 and $15.

Is GPT-5.4 or Qwen3.7 Max better for coding?

GPT-5.4 scores higher on coding benchmarks: 52.6 versus 50.4 in the Noometry coding category.

Which has the bigger context window?

GPT-5.4 does, with 1.05M tokens against 1M.

How many benchmarks do GPT-5.4 and Qwen3.7 Max share?

32 benchmarks have published results for both models. GPT-5.4 has 68 scored results on Noometry and Qwen3.7 Max has 33.

Related comparisons

Go deeper