Noometry

Model comparison

Grok 4.20 (Non-Reasoning) vs Qwen3 235B-A22B

Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.6 to 43.5 on the Noometry Index.

Last verified . 31 shared benchmarks.

Grok 4.20 (Non-Reasoning) xAI

48.6

Rank #54 Confirmed

Qwen3 235B-A22B Alibaba (Qwen)

43.5

Rank #91 Confirmed

Summary

  • They share 31 benchmarks with published results for both. Grok 4.20 (Non-Reasoning) scores higher in 6 categories and Qwen3 235B-A22B in 3 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Grok 4.20 (Non-Reasoning) leads 52.3 to 15.7.
  • The biggest single-benchmark swing is ARC-AGI-1: 89.5% for Grok 4.20 (Non-Reasoning) and 11% for Qwen3 235B-A22B.
  • Qwen3 235B-A22B is cheaper at $0.70 / $2.80 per million input/output tokens, against $1.25 / $2.50 for Grok 4.20 (Non-Reasoning).
  • Grok 4.20 (Non-Reasoning) accepts more context: 1M tokens versus 131K.
  • Qwen3 235B-A22B has downloadable open weights; the other is API-only.

Side by side

Grok 4.20 (Non-Reasoning)Qwen3 235B-A22B
ProviderxAIAlibaba (Qwen)
Noometry Index48.643.5
Released2026-02-172025-04
WeightsProprietaryOpen
Context window1M131K
Max output30K16K
Input $ / M tokens$1.25$0.70
Output $ / M tokens$2.50$2.80
Results tracked4649

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

Category by category

Coding Qwen3 235B-A22B leads

Grok 4.20 (Non-Reasoning): 42.1 (#112), Qwen3 235B-A22B: 44.3 (#75)

BenchmarkGrok 4.20 (Non-Reasoning)Qwen3 235B-A22B
WeirdML52.3%41%
LMArena Coding14591445
Aider Polyglot—59.6%
LMArena WebDev1375—
SciCode—42.4%
ALE-Bench1,150—

Agentic & Tool Use Too close to call

Grok 4.20 (Non-Reasoning): 34.4 (#46), Qwen3 235B-A22B: 33.9 (#51)

BenchmarkGrok 4.20 (Non-Reasoning)Qwen3 235B-A22B
Vending-Bench 24,663-11.34
Terminal-Bench57.3%—
Berkeley Function Calling Leaderboard—52.1%
τ²-bench Banking18%—
LMArena Search1189—

Reasoning Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 52.3 (#32), Qwen3 235B-A22B: 15.7 (#311)

BenchmarkGrok 4.20 (Non-Reasoning)Qwen3 235B-A22B
ARC-AGI-265.1%1.3%
Kagi LLM Benchmark75%69.4%
ARC-AGI-189.5%11%
Chess Puzzles24%12%
LMArena Hard Prompts14511433
DTBench90.1%80.3%
LMCA38.7%29.3%
Epoch Capabilities Index151.98143.85
ForecastBench61.459.7
SimpleBench—31%
NYT Connections (extended)85.4%—
CritPt—0%
Thematic Generalization63.8%—
Mystery Game Puzzles—9%

Math Qwen3 235B-A22B leads

Grok 4.20 (Non-Reasoning): 48.2 (#65), Qwen3 235B-A22B: 50.4 (#57)

BenchmarkGrok 4.20 (Non-Reasoning)Qwen3 235B-A22B
OTIS Mock AIME 2024-202592.2%86.7%
LMArena Math14551432
FrontierMath (Tiers 1-3)44.9%—
FrontierMath Tier 417.1%—
ProofBench14%—
Omni-MATH—71.8%
MATH Level 5—68.9%
FrontierMath (Feb 2025 set)—8.5%
FrontierMath Tier 4 (v1)—0%

Knowledge Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 52.8 (#60), Qwen3 235B-A22B: 49.6 (#73)

BenchmarkGrok 4.20 (Non-Reasoning)Qwen3 235B-A22B
GPQA Diamond89.3%80.1%
SimpleQA Verified30.2%40.4%
LMArena Expert14391463
MMLU-Pro—84.4%
Confabulations—15.6%
Vectara Hallucination Rate—9.3%
GPQA (HELM)—72.7%

Multimodal Not comparable

Grok 4.20 (Non-Reasoning): 33.3 (#98), Qwen3 235B-A22B: —

BenchmarkGrok 4.20 (Non-Reasoning)Qwen3 235B-A22B
LMArena Vision1263—
Blueprint-Bench 20%—
LMArena Document1416—

Multilingual Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 54.5 (#40), Qwen3 235B-A22B: 52.3 (#89)

BenchmarkGrok 4.20 (Non-Reasoning)Qwen3 235B-A22B
LMArena Non-English14411409
LMArena Chinese14811481
LMArena French14761445
LMArena German14651433
LMArena Japanese14491399
LMArena Korean14171391
LMArena Russian14581411
LMArena Spanish14431430

Instruction Following Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 74.8 (#83), Qwen3 235B-A22B: 72.6 (#136)

BenchmarkGrok 4.20 (Non-Reasoning)Qwen3 235B-A22B
LMArena Instruction Following14201408
IFEval—83.5%

Long Context Too close to call

Grok 4.20 (Non-Reasoning): 45.5 (#34), Qwen3 235B-A22B: 46.1 (#26)

BenchmarkGrok 4.20 (Non-Reasoning)Qwen3 235B-A22B
LMArena Longer Query14371426
Fiction.LiveBench—75%
CL-bench22.2%—
CL-bench Life11.9%—

Writing & Preference Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 65.7 (#44), Qwen3 235B-A22B: 59.6 (#108)

BenchmarkGrok 4.20 (Non-Reasoning)Qwen3 235B-A22B
LMArena Text14511419
LMArena Creative Writing14381384
EQ-Bench Creative Writing15741366
LMArena Multi-Turn14561432
Short-Story Creative Writing—83%
WildBench—86.6%

Frequently asked questions

Is Grok 4.20 (Non-Reasoning) better than Qwen3 235B-A22B?

Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.6 to 43.5 on the Noometry Index.

Which is cheaper, Grok 4.20 (Non-Reasoning) 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; Grok 4.20 (Non-Reasoning) lists at $1.25 and $2.50.

Is Grok 4.20 (Non-Reasoning) or Qwen3 235B-A22B better for coding?

Qwen3 235B-A22B scores higher on coding benchmarks: 44.3 versus 42.1 in the Noometry coding category.

Which has the bigger context window?

Grok 4.20 (Non-Reasoning) does, with 1M tokens against 131K.

How many benchmarks do Grok 4.20 (Non-Reasoning) and Qwen3 235B-A22B share?

31 benchmarks have published results for both models. Grok 4.20 (Non-Reasoning) has 46 scored results on Noometry and Qwen3 235B-A22B has 49.

Related comparisons

Go deeper