Noometry

Model comparison

Grok 4.20 (Non-Reasoning) vs Qwen3-4B

Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.5 to 31.5 on the Noometry Index.

Last verified . 3 shared benchmarks.

Grok 4.20 (Non-Reasoning) xAI

48.5

Rank #48 Confirmed

Qwen3-4B Alibaba (Qwen)

31.5

Rank #261 Confirmed

Summary

  • They share 3 benchmarks with published results for both. Grok 4.20 (Non-Reasoning) scores higher in 4 categories and Qwen3-4B in 0 categories; 4 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Grok 4.20 (Non-Reasoning) leads 52.2 to 18.5.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 92.2% for Grok 4.20 (Non-Reasoning) and 52.2% for Qwen3-4B.
  • Qwen3-4B has downloadable open weights; the other is API-only.

Side by side

Grok 4.20 (Non-Reasoning)Qwen3-4B
ProviderxAIAlibaba (Qwen)
Noometry Index48.531.5
Released2026-02-172025-04-29
WeightsProprietaryOpen
Context window1M—
Max output30K—
Input $ / M tokens$1.25—
Output $ / M tokens$2.50—
Results tracked466

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

Category by category

Coding Not comparable

Grok 4.20 (Non-Reasoning): 41.8 (#104), Qwen3-4B: —

BenchmarkGrok 4.20 (Non-Reasoning)Qwen3-4B
LMArena WebDev1374—
WeirdML52.3%—
LMArena Coding1458—
ALE-Bench1,150—

Agentic & Tool Use Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 34.3 (#43), Qwen3-4B: 27.3 (#94)

BenchmarkGrok 4.20 (Non-Reasoning)Qwen3-4B
Terminal-Bench57.3%—
Berkeley Function Calling Leaderboard—35.7%
τ²-bench Banking18%—
LMArena Search1189—
Vending-Bench 24,663—

Reasoning Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 52.2 (#30), Qwen3-4B: 18.5 (#264)

BenchmarkGrok 4.20 (Non-Reasoning)Qwen3-4B
Chess Puzzles24%4%
ARC-AGI-265.1%—
Kagi LLM Benchmark75%—
NYT Connections (extended)85.4%—
ARC-AGI-189.5%—
Thematic Generalization63.8%—
LMArena Hard Prompts1451—
DTBench90.1%—
LMCA38.7%—
Epoch Capabilities Index152.01—
ForecastBench61.4—

Math Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 48.0 (#54), Qwen3-4B: 28.5 (#228)

BenchmarkGrok 4.20 (Non-Reasoning)Qwen3-4B
OTIS Mock AIME 2024-202592.2%52.2%
FrontierMath (Tiers 1-3)44.9%—
FrontierMath Tier 417.1%—
MathArena Final-Answer Competitions—38.5%
ProofBench14%—
LMArena Math1455—

Knowledge Grok 4.20 (Non-Reasoning) leads

Grok 4.20 (Non-Reasoning): 52.8 (#54), Qwen3-4B: 33.0 (#198)

BenchmarkGrok 4.20 (Non-Reasoning)Qwen3-4B
GPQA Diamond89.3%52.3%
SimpleQA Verified30.2%—
Vectara Hallucination Rate—5.7%
LMArena Expert1440—

Multimodal Not comparable

Grok 4.20 (Non-Reasoning): 33.2 (#88), Qwen3-4B: —

BenchmarkGrok 4.20 (Non-Reasoning)Qwen3-4B
LMArena Vision1263—
Blueprint-Bench 20%—
LMArena Document1416—

Multilingual Not comparable

Grok 4.20 (Non-Reasoning): 54.7 (#38), Qwen3-4B: —

BenchmarkGrok 4.20 (Non-Reasoning)Qwen3-4B
LMArena Non-English1440—
LMArena Chinese1481—
LMArena French1477—
LMArena German1463—
LMArena Japanese1450—
LMArena Korean1417—
LMArena Russian1457—
LMArena Spanish1442—

Instruction Following Not comparable

Grok 4.20 (Non-Reasoning): 75.0 (#79), Qwen3-4B: —

BenchmarkGrok 4.20 (Non-Reasoning)Qwen3-4B
LMArena Instruction Following1420—

Long Context Not comparable

Grok 4.20 (Non-Reasoning): 45.4 (#33), Qwen3-4B: —

BenchmarkGrok 4.20 (Non-Reasoning)Qwen3-4B
CL-bench22.2%—
CL-bench Life11.9%—
LMArena Longer Query1436—

Writing & Preference Not comparable

Grok 4.20 (Non-Reasoning): 65.9 (#42), Qwen3-4B: —

BenchmarkGrok 4.20 (Non-Reasoning)Qwen3-4B
LMArena Text1451—
LMArena Creative Writing1439—
EQ-Bench Creative Writing1574—
LMArena Multi-Turn1456—

Frequently asked questions

Is Grok 4.20 (Non-Reasoning) better than Qwen3-4B?

Grok 4.20 (Non-Reasoning) is the stronger model overall, scoring 48.5 to 31.5 on the Noometry Index.

How many benchmarks do Grok 4.20 (Non-Reasoning) and Qwen3-4B share?

3 benchmarks have published results for both models. Grok 4.20 (Non-Reasoning) has 46 scored results on Noometry and Qwen3-4B has 6.

Related comparisons

Go deeper