Noometry

Model comparison

GPT-4.1 mini vs Grok 4

Grok 4 is the stronger model overall, scoring 48.1 to 33.6 on the Noometry Index.

Last verified . 35 shared benchmarks.

GPT-4.1 mini OpenAI

33.6

Rank #240 Confirmed

Grok 4 xAI

48.1

Rank #56 Confirmed

Summary

  • They share 35 benchmarks with published results for both. GPT-4.1 mini scores higher in 2 categories and Grok 4 in 8 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in long context, where Grok 4 leads 63.1 to 31.8.
  • The biggest single-benchmark swing is ARC-AGI-1: 3.5% for GPT-4.1 mini and 66.7% for Grok 4.

Side by side

GPT-4.1 miniGrok 4
ProviderOpenAIxAI
Noometry Index33.648.1
Released2025-04-142025-07-09
WeightsProprietaryProprietary
Context window1.05M—
Max output33K—
Input $ / M tokens$0.40—
Output $ / M tokens$1.60—
Results tracked4748

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

Category by category

Coding Grok 4 leads

GPT-4.1 mini: 30.6 (#293), Grok 4: 50.3 (#46)

BenchmarkGPT-4.1 miniGrok 4
Aider Polyglot32.4%79.6%
WeirdML37.6%45.7%
LMArena Coding13671408
SWE-bench Verified (bash only)23.9%—
SciCode40.4%—
BigCodeBench Instruct48.9%—
CadEval16%—

Agentic & Tool Use Too close to call

GPT-4.1 mini: 33.3 (#55), Grok 4: 32.3 (#68)

BenchmarkGPT-4.1 miniGrok 4
Berkeley Function Calling Leaderboard50.5%63%
Terminal-Bench—27.2%
GDPval—21.1%
Cybench—43%
DeepResearch Bench—47.3%
BALROG—43.6%
LMArena Search—1142
METR Time Horizons—66.6%

Reasoning Grok 4 leads

GPT-4.1 mini: 10.8 (#340), Grok 4: 36.7 (#65)

BenchmarkGPT-4.1 miniGrok 4
ARC-AGI-20%16%
Kagi LLM Benchmark48.6%73.6%
ARC-AGI-13.5%66.7%
Chess Puzzles7%28%
LMArena Hard Prompts13491409
Epoch Capabilities Index135.01146.44
SimpleBench—60.5%
CritPt0%—
Mystery Game Puzzles7%—
DTBench68.8%—
LMCA21.1%—
ForecastBench—60.9

Math Grok 4 leads

GPT-4.1 mini: 24.1 (#270), Grok 4: 48.4 (#64)

BenchmarkGPT-4.1 miniGrok 4
OTIS Mock AIME 2024-202544.7%84%
Omni-MATH49.1%60.3%
LMArena Math13431422
FrontierMath (Feb 2025 set)4.5%19.7%
FrontierMath (Tiers 1-3)6.7%—
MATH Level 587.3%—
FrontierMath Tier 4 (v1)—2.1%

Knowledge Grok 4 leads

GPT-4.1 mini: 34.7 (#194), Grok 4: 53.8 (#55)

BenchmarkGPT-4.1 miniGrok 4
GPQA Diamond65.8%87%
MMLU-Pro78.3%85.1%
GPQA (HELM)61.4%72.7%
LMArena Expert13381415
SimpleQA Verified12.7%—
Confabulations—12.4%

Multimodal GPT-4.1 mini leads

GPT-4.1 mini: 35.8 (#82), Grok 4: 33.7 (#94)

BenchmarkGPT-4.1 miniGrok 4
LMArena Vision11811210
GeoBench—45%

Multilingual Grok 4 leads

GPT-4.1 mini: 45.7 (#166), Grok 4: 51.8 (#103)

BenchmarkGPT-4.1 miniGrok 4
LMArena Non-English13181403
LMArena Chinese13291427
LMArena French13581418
LMArena German13511429
LMArena Japanese12901394
LMArena Korean12981377
LMArena Russian13241410
LMArena Spanish13191420

Instruction Following Grok 4 leads

GPT-4.1 mini: 73.7 (#118), Grok 4: 79.2 (#5)

BenchmarkGPT-4.1 miniGrok 4
IFEval90.4%94.9%
LMArena Instruction Following13331387

Long Context Grok 4 leads

GPT-4.1 mini: 31.8 (#275), Grok 4: 63.1 (#4)

BenchmarkGPT-4.1 miniGrok 4
Fiction.LiveBench44.4%94.4%
LMArena Longer Query13441409

Writing & Preference Grok 4 leads

GPT-4.1 mini: 48.6 (#199), Grok 4: 58.5 (#116)

BenchmarkGPT-4.1 miniGrok 4
LMArena Text13401411
LMArena Creative Writing13001397
WildBench83.8%79.7%
LMArena Multi-Turn13541416
Short-Story Creative Writing—76.9%
EQ-Bench Creative Writing1147—

Frequently asked questions

Is GPT-4.1 mini better than Grok 4?

Grok 4 is the stronger model overall, scoring 48.1 to 33.6 on the Noometry Index.

Is GPT-4.1 mini or Grok 4 better for coding?

Grok 4 scores higher on coding benchmarks: 50.3 versus 30.6 in the Noometry coding category.

How many benchmarks do GPT-4.1 mini and Grok 4 share?

35 benchmarks have published results for both models. GPT-4.1 mini has 47 scored results on Noometry and Grok 4 has 48.

Related comparisons

Go deeper