Noometry

Model comparison

Grok 4 vs Qwen3 235B-A22B

Grok 4 is the stronger model overall, scoring 48.1 to 43.5 on the Noometry Index.

Last verified . 39 shared benchmarks.

Grok 4 xAI

48.1

Rank #56 Confirmed

Qwen3 235B-A22B Alibaba (Qwen)

43.5

Rank #91 Confirmed

Summary

  • They share 39 benchmarks with published results for both. Grok 4 scores higher in 5 categories and Qwen3 235B-A22B in 4 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Grok 4 leads 36.7 to 15.7.
  • The biggest single-benchmark swing is ARC-AGI-1: 66.7% for Grok 4 and 11% for Qwen3 235B-A22B.
  • Qwen3 235B-A22B has downloadable open weights; the other is API-only.

Side by side

Grok 4Qwen3 235B-A22B
ProviderxAIAlibaba (Qwen)
Noometry Index48.143.5
Released2025-07-092025-04
WeightsProprietaryOpen
Context window—131K
Max output—16K
Input $ / M tokens—$0.70
Output $ / M tokens—$2.80
Results tracked4849

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

Category by category

Coding Grok 4 leads

Grok 4: 50.3 (#46), Qwen3 235B-A22B: 44.3 (#75)

BenchmarkGrok 4Qwen3 235B-A22B
Aider Polyglot79.6%59.6%
WeirdML45.7%41%
LMArena Coding14081445
SciCode—42.4%

Agentic & Tool Use Qwen3 235B-A22B leads

Grok 4: 32.3 (#68), Qwen3 235B-A22B: 33.9 (#51)

BenchmarkGrok 4Qwen3 235B-A22B
Berkeley Function Calling Leaderboard63%52.1%
Terminal-Bench27.2%—
GDPval21.1%—
Cybench43%—
DeepResearch Bench47.3%—
BALROG43.6%—
LMArena Search1142—
METR Time Horizons66.6%—
Vending-Bench 2—-11.34

Reasoning Grok 4 leads

Grok 4: 36.7 (#65), Qwen3 235B-A22B: 15.7 (#311)

BenchmarkGrok 4Qwen3 235B-A22B
ARC-AGI-216%1.3%
SimpleBench60.5%31%
Kagi LLM Benchmark73.6%69.4%
ARC-AGI-166.7%11%
Chess Puzzles28%12%
LMArena Hard Prompts14091433
Epoch Capabilities Index146.44143.85
ForecastBench60.959.7
CritPt—0%
Mystery Game Puzzles—9%
DTBench—80.3%
LMCA—29.3%

Math Qwen3 235B-A22B leads

Grok 4: 48.4 (#64), Qwen3 235B-A22B: 50.4 (#57)

BenchmarkGrok 4Qwen3 235B-A22B
OTIS Mock AIME 2024-202584%86.7%
Omni-MATH60.3%71.8%
LMArena Math14221432
FrontierMath (Feb 2025 set)19.7%8.5%
FrontierMath Tier 4 (v1)2.1%0%
MATH Level 5—68.9%

Knowledge Grok 4 leads

Grok 4: 53.8 (#55), Qwen3 235B-A22B: 49.6 (#73)

BenchmarkGrok 4Qwen3 235B-A22B
GPQA Diamond87%80.1%
MMLU-Pro85.1%84.4%
Confabulations12.4%15.6%
GPQA (HELM)72.7%72.7%
LMArena Expert14151463
SimpleQA Verified—40.4%
Vectara Hallucination Rate—9.3%

Multimodal Not comparable

Grok 4: 33.7 (#94), Qwen3 235B-A22B: —

BenchmarkGrok 4Qwen3 235B-A22B
LMArena Vision1210—
GeoBench45%—

Multilingual Too close to call

Grok 4: 51.8 (#103), Qwen3 235B-A22B: 52.3 (#89)

BenchmarkGrok 4Qwen3 235B-A22B
LMArena Non-English14031409
LMArena Chinese14271481
LMArena French14181445
LMArena German14291433
LMArena Japanese13941399
LMArena Korean13771391
LMArena Russian14101411
LMArena Spanish14201430

Instruction Following Grok 4 leads

Grok 4: 79.2 (#5), Qwen3 235B-A22B: 72.6 (#136)

BenchmarkGrok 4Qwen3 235B-A22B
IFEval94.9%83.5%
LMArena Instruction Following13871408

Long Context Grok 4 leads

Grok 4: 63.1 (#4), Qwen3 235B-A22B: 46.1 (#26)

BenchmarkGrok 4Qwen3 235B-A22B
Fiction.LiveBench94.4%75%
LMArena Longer Query14091426

Writing & Preference Qwen3 235B-A22B leads

Grok 4: 58.5 (#116), Qwen3 235B-A22B: 59.6 (#108)

BenchmarkGrok 4Qwen3 235B-A22B
LMArena Text14111419
LMArena Creative Writing13971384
Short-Story Creative Writing76.9%83%
WildBench79.7%86.6%
LMArena Multi-Turn14161432
EQ-Bench Creative Writing—1366

Frequently asked questions

Is Grok 4 better than Qwen3 235B-A22B?

Grok 4 is the stronger model overall, scoring 48.1 to 43.5 on the Noometry Index.

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

Grok 4 scores higher on coding benchmarks: 50.3 versus 44.3 in the Noometry coding category.

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

39 benchmarks have published results for both models. Grok 4 has 48 scored results on Noometry and Qwen3 235B-A22B has 49.

Related comparisons

Go deeper