Noometry

Model comparison

Grok 4 vs Qwen3.5 27B

Grok 4 is the stronger model overall, scoring 48.1 to 41.9 on the Noometry Index.

Last verified . 19 shared benchmarks.

Grok 4 xAI

48.1

Rank #56 Confirmed

Qwen3.5 27B Alibaba (Qwen)

41.9

Rank #127 Confirmed

Summary

  • They share 19 benchmarks with published results for both. Grok 4 scores higher in 7 categories and Qwen3.5 27B in 2 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in long context, where Grok 4 leads 63.1 to 43.1.
  • The biggest single-benchmark swing is WeirdML: 45.7% for Grok 4 and 39.5% for Qwen3.5 27B.
  • Qwen3.5 27B has downloadable open weights; the other is API-only.

Side by side

Grok 4Qwen3.5 27B
ProviderxAIAlibaba (Qwen)
Noometry Index48.141.9
Released2025-07-092026-02-23
WeightsProprietaryOpen
Context window—262K
Max output—66K
Input $ / M tokens—$0.30
Output $ / M tokens—$2.40
Results tracked4828

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

Category by category

Coding Grok 4 leads

Grok 4: 50.3 (#46), Qwen3.5 27B: 38.9 (#168)

BenchmarkGrok 4Qwen3.5 27B
WeirdML45.7%39.5%
LMArena Coding14081427
Aider Polyglot79.6%—
LMArena WebDev—1358
ALE-Bench—349.45

Agentic & Tool Use Not comparable

Grok 4: 32.3 (#68), Qwen3.5 27B: —

BenchmarkGrok 4Qwen3.5 27B
Terminal-Bench27.2%—
Berkeley Function Calling Leaderboard63%—
GDPval21.1%—
Cybench43%—
DeepResearch Bench47.3%—
BALROG43.6%—
LMArena Search1142—
METR Time Horizons66.6%—
Vending-Bench 2—201.98

Reasoning Grok 4 leads

Grok 4: 36.7 (#65), Qwen3.5 27B: 27.5 (#117)

BenchmarkGrok 4Qwen3.5 27B
LMArena Hard Prompts14091414
ARC-AGI-216%—
SimpleBench60.5%—
Kagi LLM Benchmark73.6%—
NYT Connections (extended)—47.9%
ARC-AGI-166.7%—
Chess Puzzles28%—
Thematic Generalization—45.5%
DTBench—82.4%
LMCA—34%
Epoch Capabilities Index146.44—
ForecastBench60.9—

Math Grok 4 leads

Grok 4: 48.4 (#64), Qwen3.5 27B: 38.8 (#127)

BenchmarkGrok 4Qwen3.5 27B
LMArena Math14221429
MathArena Final-Answer Competitions—56.7%
OTIS Mock AIME 2024-202584%—
Omni-MATH60.3%—
FrontierMath (Feb 2025 set)19.7%—
FrontierMath Tier 4 (v1)2.1%—

Knowledge Grok 4 leads

Grok 4: 53.8 (#55), Qwen3.5 27B: 38.0 (#150)

BenchmarkGrok 4Qwen3.5 27B
LMArena Expert14151428
GPQA Diamond87%—
MMLU-Pro85.1%—
Confabulations12.4%—
Vectara Hallucination Rate—12.1%
GPQA (HELM)72.7%—

Multimodal Qwen3.5 27B leads

Grok 4: 33.7 (#94), Qwen3.5 27B: 39.4 (#59)

BenchmarkGrok 4Qwen3.5 27B
LMArena Vision12101241
GeoBench45%—

Multilingual Too close to call

Grok 4: 51.8 (#103), Qwen3.5 27B: 50.8 (#115)

BenchmarkGrok 4Qwen3.5 27B
LMArena Non-English14031390
LMArena Chinese14271478
LMArena French14181410
LMArena German14291393
LMArena Japanese13941345
LMArena Korean13771358
LMArena Russian14101390
LMArena Spanish14201407

Instruction Following Grok 4 leads

Grok 4: 79.2 (#5), Qwen3.5 27B: 73.5 (#119)

BenchmarkGrok 4Qwen3.5 27B
LMArena Instruction Following13871393
IFEval94.9%—

Long Context Grok 4 leads

Grok 4: 63.1 (#4), Qwen3.5 27B: 43.1 (#106)

BenchmarkGrok 4Qwen3.5 27B
LMArena Longer Query14091413
Fiction.LiveBench94.4%—

Writing & Preference Too close to call

Grok 4: 58.5 (#116), Qwen3.5 27B: 59.3 (#111)

BenchmarkGrok 4Qwen3.5 27B
LMArena Text14111409
LMArena Creative Writing13971362
LMArena Multi-Turn14161410
Short-Story Creative Writing76.9%—
WildBench79.7%—

Frequently asked questions

Is Grok 4 better than Qwen3.5 27B?

Grok 4 is the stronger model overall, scoring 48.1 to 41.9 on the Noometry Index.

Is Grok 4 or Qwen3.5 27B better for coding?

Grok 4 scores higher on coding benchmarks: 50.3 versus 38.9 in the Noometry coding category.

How many benchmarks do Grok 4 and Qwen3.5 27B share?

19 benchmarks have published results for both models. Grok 4 has 48 scored results on Noometry and Qwen3.5 27B has 28.

Related comparisons

Go deeper