Noometry

Model comparison

Grok 4 vs Grok 4.5

Grok 4.5 is the stronger model overall, scoring 55.0 to 48.1 on the Noometry Index.

Last verified . 28 shared benchmarks.

Grok 4 xAI

48.1

Rank #56 Confirmed

Grok 4.5 xAI

55.0

Rank #25 Confirmed

Summary

  • They share 28 benchmarks with published results for both. Grok 4 scores higher in 2 categories and Grok 4.5 in 8 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Grok 4.5 leads 56.1 to 36.7.
  • The biggest single-benchmark swing is ARC-AGI-2: 16% for Grok 4 and 52.6% for Grok 4.5.

Side by side

Grok 4Grok 4.5
ProviderxAIxAI
Noometry Index48.155.0
Released2025-07-092026-07-08
WeightsProprietaryProprietary
Context window—500K
Max output—500K
Input $ / M tokens—$2
Output $ / M tokens—$6
Results tracked4852

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

Category by category

Coding Grok 4.5 leads

Grok 4: 50.3 (#46), Grok 4.5: 52.2 (#35)

BenchmarkGrok 4Grok 4.5
WeirdML45.7%46.4%
LMArena Coding14081474
DeepSWE—53.8%
FrontierCode—42.4%
Aider Polyglot79.6%—
LMArena WebDev—1553
SciCode—54.1%
ALE-Bench—1,309

Agentic & Tool Use Grok 4.5 leads

Grok 4: 32.3 (#68), Grok 4.5: 44.4 (#17)

BenchmarkGrok 4Grok 4.5
LMArena Search11421213
Terminal-Bench27.2%—
APEX-Agents—56.2%
Berkeley Function Calling Leaderboard63%—
GDPval21.1%—
τ²-bench Banking—47.9%
Cybench43%—
DeepResearch Bench47.3%—
PostTrainBench—23.4%
BALROG43.6%—
GBAEval—65.4%
GDP.pdf—14%
METR Time Horizons66.6%—
Vending-Bench 2—3,887

Reasoning Grok 4.5 leads

Grok 4: 36.7 (#65), Grok 4.5: 56.1 (#25)

BenchmarkGrok 4Grok 4.5
ARC-AGI-216%52.6%
SimpleBench60.5%70%
Kagi LLM Benchmark73.6%83.5%
ARC-AGI-166.7%87.2%
Chess Puzzles28%36%
LMArena Hard Prompts14091462
Epoch Capabilities Index146.44153.92
NYT Connections (extended)—79.9%
CritPt—15.4%
DTBench—96.5%
LMCA—45.2%
Surface Evolver Bench—74.4%
ForecastBench60.9—

Math Grok 4.5 leads

Grok 4: 48.4 (#64), Grok 4.5: 60.9 (#35)

BenchmarkGrok 4Grok 4.5
OTIS Mock AIME 2024-202584%97.8%
LMArena Math14221459
FrontierMath (Tiers 1-3)—57.2%
FrontierMath Tier 4—24.4%
ProofBench—31%
Omni-MATH60.3%—
FrontierMath (Feb 2025 set)19.7%—
FrontierMath Tier 4 (v1)2.1%—

Knowledge Grok 4.5 leads

Grok 4: 53.8 (#55), Grok 4.5: 62.3 (#24)

BenchmarkGrok 4Grok 4.5
GPQA Diamond87%93.4%
LMArena Expert14151466
SimpleQA Verified—48.3%
MMLU-Pro85.1%—
Confabulations12.4%—
GPQA (HELM)72.7%—

Multimodal Grok 4.5 leads

Grok 4: 33.7 (#94), Grok 4.5: 37.6 (#72)

BenchmarkGrok 4Grok 4.5
LMArena Vision12101288
GeoBench45%—
Blueprint-Bench 2—27.3%
Furniture Assembly—22.5%
LMArena Document—1452

Multilingual Grok 4.5 leads

Grok 4: 51.8 (#103), Grok 4.5: 54.4 (#42)

BenchmarkGrok 4Grok 4.5
LMArena Non-English14031440
LMArena Chinese14271496
LMArena French14181456
LMArena German14291446
LMArena Japanese13941428
LMArena Korean13771404
LMArena Russian14101448
LMArena Spanish14201450

Instruction Following Grok 4 leads

Grok 4: 79.2 (#5), Grok 4.5: 76.0 (#48)

BenchmarkGrok 4Grok 4.5
LMArena Instruction Following13871446
IFEval94.9%—

Long Context Grok 4 leads

Grok 4: 63.1 (#4), Grok 4.5: 44.8 (#56)

BenchmarkGrok 4Grok 4.5
LMArena Longer Query14091463
Fiction.LiveBench94.4%—

Writing & Preference Grok 4.5 leads

Grok 4: 58.5 (#116), Grok 4.5: 65.8 (#42)

BenchmarkGrok 4Grok 4.5
LMArena Text14111448
LMArena Creative Writing13971442
LMArena Multi-Turn14161456
Short-Story Creative Writing76.9%—
EQ-Bench Creative Writing—1579
WildBench79.7%—

Frequently asked questions

Is Grok 4 better than Grok 4.5?

Grok 4.5 is the stronger model overall, scoring 55.0 to 48.1 on the Noometry Index.

Is Grok 4 or Grok 4.5 better for coding?

Grok 4.5 scores higher on coding benchmarks: 52.2 versus 50.3 in the Noometry coding category.

How many benchmarks do Grok 4 and Grok 4.5 share?

28 benchmarks have published results for both models. Grok 4 has 48 scored results on Noometry and Grok 4.5 has 52.

Related comparisons

Go deeper