Noometry

Model comparison

Grok 4 vs Qwen3.6 35B-A3B

Grok 4 is the stronger model overall, scoring 47.8 to 37.2 on the Noometry Index.

Last verified . 6 shared benchmarks.

Grok 4 xAI

47.8

Rank #49 Confirmed

Qwen3.6 35B-A3B Alibaba (Qwen)

37.2

Rank #194 Confirmed

Summary

  • They share 6 benchmarks with published results for both. Grok 4 scores higher in 5 categories and Qwen3.6 35B-A3B in 0 categories; 5 gaps are clear of the uncertainty.
  • The widest gap is in coding, where Grok 4 leads 49.9 to 36.7.
  • The biggest single-benchmark swing is WeirdML: 45.7% for Grok 4 and 34.5% for Qwen3.6 35B-A3B.
  • Qwen3.6 35B-A3B has downloadable open weights; the other is API-only.

Side by side

Grok 4Qwen3.6 35B-A3B
ProviderxAIAlibaba (Qwen)
Noometry Index47.837.2
Released2025-07-092026-04-01
WeightsProprietaryOpen
Context window—262K
Max output—66K
Input $ / M tokens—$0.25
Output $ / M tokens—$1.49
Results tracked4814

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

Category by category

Coding Grok 4 leads

Grok 4: 49.9 (#44), Qwen3.6 35B-A3B: 36.7 (#190)

BenchmarkGrok 4Qwen3.6 35B-A3B
WeirdML45.7%34.5%
Aider Polyglot79.6%—
SciCode—35.8%
LMArena Coding1408—

Agentic & Tool Use Grok 4 leads

Grok 4: 32.1 (#62), Qwen3.6 35B-A3B: 21.9 (#126)

BenchmarkGrok 4Qwen3.6 35B-A3B
Terminal-Bench27.2%23%
Berkeley Function Calling Leaderboard63%—
GDPval21.1%—
Cybench43%—
DeepResearch Bench47.3%—
BALROG43.6%—
LMArena Search1142—
METR Time Horizons66.6%—

Reasoning Grok 4 leads

Grok 4: 36.9 (#59), Qwen3.6 35B-A3B: 27.2 (#112)

BenchmarkGrok 4Qwen3.6 35B-A3B
Chess Puzzles28%26%
Epoch Capabilities Index146.46143.92
ARC-AGI-216%—
SimpleBench60.5%—
Kagi LLM Benchmark73.6%—
NYT Connections (extended)—41.6%
ARC-AGI-166.7%—
CritPt—0.3%
LMArena Hard Prompts1409—
Mystery Game Puzzles—22%
DTBench—73.9%
LMCA—29.7%
Surface Evolver Bench—44.4%
ForecastBench60.9—

Math Grok 4 leads

Grok 4: 46.5 (#58), Qwen3.6 35B-A3B: 37.9 (#90)

BenchmarkGrok 4Qwen3.6 35B-A3B
OTIS Mock AIME 2024-202584%86.7%
FrontierMath (Tiers 1-3)—20.4%
Omni-MATH60.3%—
LMArena Math1422—
FrontierMath (Feb 2025 set)19.7%—
FrontierMath Tier 4 (v1)2.1%—

Knowledge Grok 4 leads

Grok 4: 53.7 (#50), Qwen3.6 35B-A3B: 51.4 (#63)

BenchmarkGrok 4Qwen3.6 35B-A3B
GPQA Diamond87%84.8%
MMLU-Pro85.1%—
Confabulations12.4%—
GPQA (HELM)72.7%—
LMArena Expert1417—

Multimodal Not comparable

Grok 4: 33.1 (#89), Qwen3.6 35B-A3B: —

BenchmarkGrok 4Qwen3.6 35B-A3B
LMArena Vision1211—
GeoBench45%—

Multilingual Not comparable

Grok 4: 52.0 (#98), Qwen3.6 35B-A3B: —

BenchmarkGrok 4Qwen3.6 35B-A3B
LMArena Non-English1403—
LMArena Chinese1428—
LMArena French1418—
LMArena German1429—
LMArena Japanese1396—
LMArena Korean1376—
LMArena Russian1409—
LMArena Spanish1419—

Instruction Following Not comparable

Grok 4: 79.4 (#4), Qwen3.6 35B-A3B: —

BenchmarkGrok 4Qwen3.6 35B-A3B
IFEval94.9%—
LMArena Instruction Following1387—

Long Context Not comparable

Grok 4: 63.1 (#4), Qwen3.6 35B-A3B: —

BenchmarkGrok 4Qwen3.6 35B-A3B
Fiction.LiveBench94.4%—
LMArena Longer Query1409—

Writing & Preference Not comparable

Grok 4: 58.7 (#111), Qwen3.6 35B-A3B: —

BenchmarkGrok 4Qwen3.6 35B-A3B
LMArena Text1412—
LMArena Creative Writing1397—
Short-Story Creative Writing76.9%—
WildBench79.7%—
LMArena Multi-Turn1416—

Frequently asked questions

Is Grok 4 better than Qwen3.6 35B-A3B?

Grok 4 is the stronger model overall, scoring 47.8 to 37.2 on the Noometry Index.

Is Grok 4 or Qwen3.6 35B-A3B better for coding?

Grok 4 scores higher on coding benchmarks: 49.9 versus 36.7 in the Noometry coding category.

How many benchmarks do Grok 4 and Qwen3.6 35B-A3B share?

6 benchmarks have published results for both models. Grok 4 has 48 scored results on Noometry and Qwen3.6 35B-A3B has 14.

Related comparisons

Go deeper