Noometry

Model comparison

Grok 4 vs Qwen2.5-VL 72B Instruct

Grok 4 is the stronger model overall, scoring 48.1 to 29.9 on the Noometry Index.

Last verified . 3 shared benchmarks.

Grok 4 xAI

48.1

Rank #56 Confirmed

Qwen2.5-VL 72B Instruct Alibaba (Qwen)

29.9

Rank #302 Confirmed

Summary

  • They share 3 benchmarks with published results for both. Grok 4 scores higher in 3 categories and Qwen2.5-VL 72B Instruct in 0 categories; 2 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Grok 4 leads 36.7 to 20.7.
  • The biggest single-benchmark swing is Kagi LLM Benchmark: 73.6% for Grok 4 and 36% for Qwen2.5-VL 72B Instruct.
  • Qwen2.5-VL 72B Instruct has downloadable open weights; the other is API-only.

Side by side

Grok 4Qwen2.5-VL 72B Instruct
ProviderxAIAlibaba (Qwen)
Noometry Index48.129.9
Released2025-07-092024-09
WeightsProprietaryOpen
Context window—131K
Max output—8K
Input $ / M tokens—$2.80
Output $ / M tokens—$8.40
Results tracked486

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

Category by category

Coding Not comparable

Grok 4: 50.3 (#46), Qwen2.5-VL 72B Instruct: —

BenchmarkGrok 4Qwen2.5-VL 72B Instruct
Aider Polyglot79.6%—
WeirdML45.7%—
LMArena Coding1408—

Agentic & Tool Use Grok 4 leads

Grok 4: 32.3 (#68), Qwen2.5-VL 72B Instruct: 18.6 (#144)

BenchmarkGrok 4Qwen2.5-VL 72B Instruct
Terminal-Bench27.2%—
Berkeley Function Calling Leaderboard63%—
GDPval21.1%—
Cybench43%—
DeepResearch Bench47.3%—
OSWorld—5%
BALROG43.6%—
LMArena Search1142—
METR Time Horizons66.6%—

Reasoning Grok 4 leads

Grok 4: 36.7 (#65), Qwen2.5-VL 72B Instruct: 20.7 (#233)

BenchmarkGrok 4Qwen2.5-VL 72B Instruct
Kagi LLM Benchmark73.6%36%
ARC-AGI-216%—
SimpleBench60.5%—
ARC-AGI-166.7%—
Chess Puzzles28%—
LMArena Hard Prompts1409—
Epoch Capabilities Index146.44—
ForecastBench60.9—

Math Not comparable

Grok 4: 48.4 (#64), Qwen2.5-VL 72B Instruct: —

BenchmarkGrok 4Qwen2.5-VL 72B Instruct
OTIS Mock AIME 2024-202584%—
Omni-MATH60.3%—
LMArena Math1422—
FrontierMath (Feb 2025 set)19.7%—
FrontierMath Tier 4 (v1)2.1%—

Knowledge Not comparable

Grok 4: 53.8 (#55), Qwen2.5-VL 72B Instruct: —

BenchmarkGrok 4Qwen2.5-VL 72B Instruct
GPQA Diamond87%—
MMLU-Pro85.1%—
Confabulations12.4%—
GPQA (HELM)72.7%—
LMArena Expert1415—

Multimodal Too close to call

Grok 4: 33.7 (#94), Qwen2.5-VL 72B Instruct: 33.5 (#97)

BenchmarkGrok 4Qwen2.5-VL 72B Instruct
LMArena Vision12101107
GeoBench45%62%
Video-MME—73.5%
SpatialViz-Bench—33.3%

Multilingual Not comparable

Grok 4: 51.8 (#103), Qwen2.5-VL 72B Instruct: —

BenchmarkGrok 4Qwen2.5-VL 72B Instruct
LMArena Non-English1403—
LMArena Chinese1427—
LMArena French1418—
LMArena German1429—
LMArena Japanese1394—
LMArena Korean1377—
LMArena Russian1410—
LMArena Spanish1420—

Instruction Following Not comparable

Grok 4: 79.2 (#5), Qwen2.5-VL 72B Instruct: —

BenchmarkGrok 4Qwen2.5-VL 72B Instruct
IFEval94.9%—
LMArena Instruction Following1387—

Long Context Not comparable

Grok 4: 63.1 (#4), Qwen2.5-VL 72B Instruct: —

BenchmarkGrok 4Qwen2.5-VL 72B Instruct
Fiction.LiveBench94.4%—
LMArena Longer Query1409—

Writing & Preference Not comparable

Grok 4: 58.5 (#116), Qwen2.5-VL 72B Instruct: —

BenchmarkGrok 4Qwen2.5-VL 72B Instruct
LMArena Text1411—
LMArena Creative Writing1397—
Short-Story Creative Writing76.9%—
WildBench79.7%—
LMArena Multi-Turn1416—

Frequently asked questions

Is Grok 4 better than Qwen2.5-VL 72B Instruct?

Grok 4 is the stronger model overall, scoring 48.1 to 29.9 on the Noometry Index.

How many benchmarks do Grok 4 and Qwen2.5-VL 72B Instruct share?

3 benchmarks have published results for both models. Grok 4 has 48 scored results on Noometry and Qwen2.5-VL 72B Instruct has 6.

Related comparisons

Go deeper