Noometry

Model comparison

GPT-4 vs Qwen3 235B-A22B

Qwen3 235B-A22B is the stronger model overall, scoring 43.2 to 28.9 on the Noometry Index.

Last verified . 28 shared benchmarks.

GPT-4 OpenAI

28.9

Rank #308 Confirmed

Qwen3 235B-A22B Alibaba (Qwen)

43.2

Rank #83 Confirmed

Summary

  • They share 28 benchmarks with published results for both. GPT-4 scores higher in 1 category and Qwen3 235B-A22B in 7 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where Qwen3 235B-A22B leads 48.5 to 10.1.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.1% for GPT-4 and 86.7% for Qwen3 235B-A22B.
  • Qwen3 235B-A22B is cheaper at $0.70 / $2.80 per million input/output tokens, against $30 / $60 for GPT-4.
  • Qwen3 235B-A22B accepts more context: 131K tokens versus 8K.
  • Qwen3 235B-A22B has downloadable open weights; the other is API-only.

Side by side

GPT-4Qwen3 235B-A22B
ProviderOpenAIAlibaba (Qwen)
Noometry Index28.943.2
Released2023-03-142025-04
WeightsProprietaryOpen
Context window8K131K
Max output8K16K
Input $ / M tokens$30$0.70
Output $ / M tokens$60$2.80
Results tracked3849

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

Category by category

Coding Qwen3 235B-A22B leads

GPT-4: 31.2 (#277), Qwen3 235B-A22B: 43.8 (#72)

BenchmarkGPT-4Qwen3 235B-A22B
WeirdML12.4%41%
LMArena Coding12541445
Aider Polyglot—59.6%
SciCode—42.4%
BigCodeBench Instruct46%—
BigCodeBench Complete57.2%—
HumanEval+79.3%—

Agentic & Tool Use Not comparable

GPT-4: —, Qwen3 235B-A22B: 33.7 (#48)

BenchmarkGPT-4Qwen3 235B-A22B
Berkeley Function Calling Leaderboard—52.1%
METR Time Horizons36.1%—
Vending-Bench 2—-11.34

Reasoning GPT-4 leads

GPT-4: 17.3 (#285), Qwen3 235B-A22B: 15.8 (#302)

BenchmarkGPT-4Qwen3 235B-A22B
Chess Puzzles4%12%
LMArena Hard Prompts12411433
Mystery Game Puzzles12%9%
DTBench62.7%80.3%
LMCA17.1%29.3%
Epoch Capabilities Index125.88143.87
ForecastBench57.859.7
ARC-AGI-2—1.3%
SimpleBench—31%
Kagi LLM Benchmark—69.4%
ARC-AGI-1—11%
CritPt—0%
BIG-Bench Hard75.1%—
HellaSwag95.3%—
WinoGrande87.5%—

Math Qwen3 235B-A22B leads

GPT-4: 10.1 (#304), Qwen3 235B-A22B: 48.5 (#51)

BenchmarkGPT-4Qwen3 235B-A22B
OTIS Mock AIME 2024-20251.1%86.7%
LMArena Math12691431
MATH Level 523%68.9%
Omni-MATH—71.8%
FrontierMath (Feb 2025 set)—8.5%
FrontierMath Tier 4 (v1)—0%
GSM8K92%—

Knowledge Qwen3 235B-A22B leads

GPT-4: 18.4 (#277), Qwen3 235B-A22B: 49.5 (#67)

BenchmarkGPT-4Qwen3 235B-A22B
GPQA Diamond35.7%80.1%
LMArena Expert12111460
SimpleQA Verified—40.4%
MMLU-Pro—84.4%
Confabulations—15.6%
Vectara Hallucination Rate—9.3%
GPQA (HELM)—72.7%
MMLU86.4%—
TriviaQA84.8%—

Multilingual Qwen3 235B-A22B leads

GPT-4: 40.8 (#210), Qwen3 235B-A22B: 52.5 (#84)

BenchmarkGPT-4Qwen3 235B-A22B
LMArena Non-English12461410
LMArena Chinese12421481
LMArena French12821445
LMArena German12511432
LMArena Japanese12091400
LMArena Korean11841391
LMArena Russian12511411
LMArena Spanish12611429

Instruction Following Qwen3 235B-A22B leads

GPT-4: 65.5 (#217), Qwen3 235B-A22B: 72.8 (#132)

BenchmarkGPT-4Qwen3 235B-A22B
LMArena Instruction Following12411408
IFEval—83.5%

Long Context Qwen3 235B-A22B leads

GPT-4: 37.7 (#207), Qwen3 235B-A22B: 46.1 (#25)

BenchmarkGPT-4Qwen3 235B-A22B
LMArena Longer Query12441426
Fiction.LiveBench—75%

Writing & Preference Qwen3 235B-A22B leads

GPT-4: 35.0 (#263), Qwen3 235B-A22B: 59.8 (#103)

BenchmarkGPT-4Qwen3 235B-A22B
LMArena Text12631419
LMArena Creative Writing12441384
EQ-Bench Creative Writing7521366
LMArena Multi-Turn12571432
Short-Story Creative Writing—83%
WildBench—86.6%

Frequently asked questions

Is GPT-4 better than Qwen3 235B-A22B?

Qwen3 235B-A22B is the stronger model overall, scoring 43.2 to 28.9 on the Noometry Index.

Which is cheaper, GPT-4 or Qwen3 235B-A22B?

Qwen3 235B-A22B is cheaper. It lists at $0.70 per million input tokens and $2.80 per million output tokens; GPT-4 lists at $30 and $60.

Is GPT-4 or Qwen3 235B-A22B better for coding?

Qwen3 235B-A22B scores higher on coding benchmarks: 43.8 versus 31.2 in the Noometry coding category.

Which has the bigger context window?

Qwen3 235B-A22B does, with 131K tokens against 8K.

How many benchmarks do GPT-4 and Qwen3 235B-A22B share?

28 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Qwen3 235B-A22B has 49.

Related comparisons

Go deeper