Noometry

Model comparison

Grok 4 Fast vs Qwen3 235B-A22B

Qwen3 235B-A22B is the stronger model overall, scoring 43.5 to 39.4 on the Noometry Index.

Last verified . 26 shared benchmarks.

Grok 4 Fast xAI

39.4

Rank #167 Confirmed

Qwen3 235B-A22B Alibaba (Qwen)

43.5

Rank #91 Confirmed

Summary

  • They share 26 benchmarks with published results for both. Grok 4 Fast scores higher in 4 categories and Qwen3 235B-A22B in 5 categories; 6 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where Qwen3 235B-A22B leads 49.6 to 32.0.
  • The biggest single-benchmark swing is ARC-AGI-1: 48.5% for Grok 4 Fast and 11% for Qwen3 235B-A22B.
  • Qwen3 235B-A22B has downloadable open weights; the other is API-only.

Side by side

Grok 4 FastQwen3 235B-A22B
ProviderxAIAlibaba (Qwen)
Noometry Index39.443.5
Released2025-09-192025-04
WeightsProprietaryOpen
Context window—131K
Max output—16K
Input $ / M tokens—$0.70
Output $ / M tokens—$2.80
Results tracked3049

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

Category by category

Coding Qwen3 235B-A22B leads

Grok 4 Fast: 32.5 (#271), Qwen3 235B-A22B: 44.3 (#75)

BenchmarkGrok 4 FastQwen3 235B-A22B
WeirdML42.9%41%
LMArena Coding14291445
Aider Polyglot—59.6%
LMArena WebDev1159—
SciCode—42.4%

Agentic & Tool Use Qwen3 235B-A22B leads

Grok 4 Fast: 29.5 (#86), Qwen3 235B-A22B: 33.9 (#51)

BenchmarkGrok 4 FastQwen3 235B-A22B
Berkeley Function Calling Leaderboard—52.1%
τ²-bench Banking15.7%—
Cybench30%—
LMArena Search1171—
Vending-Bench 2—-11.34

Reasoning Grok 4 Fast leads

Grok 4 Fast: 22.2 (#201), Qwen3 235B-A22B: 15.7 (#311)

BenchmarkGrok 4 FastQwen3 235B-A22B
ARC-AGI-25.3%1.3%
Kagi LLM Benchmark66.1%69.4%
ARC-AGI-148.5%11%
LMArena Hard Prompts14121433
DTBench82.7%80.3%
Epoch Capabilities Index144.2143.85
ForecastBench60.559.7
SimpleBench—31%
CritPt—0%
Chess Puzzles—12%
Mystery Game Puzzles—9%
LMCA—29.3%

Math Qwen3 235B-A22B leads

Grok 4 Fast: 38.9 (#123), Qwen3 235B-A22B: 50.4 (#57)

BenchmarkGrok 4 FastQwen3 235B-A22B
LMArena Math14191432
OTIS Mock AIME 2024-2025—86.7%
Omni-MATH—71.8%
MATH Level 5—68.9%
FrontierMath (Feb 2025 set)—8.5%
FrontierMath Tier 4 (v1)—0%

Knowledge Qwen3 235B-A22B leads

Grok 4 Fast: 32.0 (#214), Qwen3 235B-A22B: 49.6 (#73)

BenchmarkGrok 4 FastQwen3 235B-A22B
Vectara Hallucination Rate19.7%9.3%
LMArena Expert14111463
GPQA Diamond—80.1%
SimpleQA Verified—40.4%
MMLU-Pro—84.4%
Confabulations—15.6%
GPQA (HELM)—72.7%

Multilingual Too close to call

Grok 4 Fast: 51.3 (#111), Qwen3 235B-A22B: 52.3 (#89)

BenchmarkGrok 4 FastQwen3 235B-A22B
LMArena Non-English13961409
LMArena Chinese14571481
LMArena French14311445
LMArena German13831433
LMArena Japanese13521399
LMArena Korean13571391
LMArena Russian13891411
LMArena Spanish14151430

Instruction Following Too close to call

Grok 4 Fast: 73.2 (#121), Qwen3 235B-A22B: 72.6 (#136)

BenchmarkGrok 4 FastQwen3 235B-A22B
LMArena Instruction Following13871408
IFEval—83.5%

Long Context Grok 4 Fast leads

Grok 4 Fast: 63.2 (#3), Qwen3 235B-A22B: 46.1 (#26)

BenchmarkGrok 4 FastQwen3 235B-A22B
Fiction.LiveBench94.4%75%
LMArena Longer Query14151426

Writing & Preference Too close to call

Grok 4 Fast: 60.0 (#102), Qwen3 235B-A22B: 59.6 (#108)

BenchmarkGrok 4 FastQwen3 235B-A22B
LMArena Text14071419
LMArena Creative Writing13871384
LMArena Multi-Turn14141432
Short-Story Creative Writing—83%
EQ-Bench Creative Writing—1366
WildBench—86.6%

Frequently asked questions

Is Grok 4 Fast better than Qwen3 235B-A22B?

Qwen3 235B-A22B is the stronger model overall, scoring 43.5 to 39.4 on the Noometry Index.

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

Qwen3 235B-A22B scores higher on coding benchmarks: 44.3 versus 32.5 in the Noometry coding category.

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

26 benchmarks have published results for both models. Grok 4 Fast has 30 scored results on Noometry and Qwen3 235B-A22B has 49.

Related comparisons

Go deeper