Noometry

Model comparison

Grok 4 vs Llama 4 Scout

Grok 4 is the stronger model overall, scoring 48.1 to 27.7 on the Noometry Index.

Last verified . 33 shared benchmarks.

Grok 4 xAI

48.1

Rank #56 Confirmed

Llama 4 Scout Meta

27.7

Rank #330 Confirmed

Summary

  • They share 33 benchmarks with published results for both. Grok 4 scores higher in 10 categories and Llama 4 Scout in 0 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in long context, where Grok 4 leads 63.1 to 27.5.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 84% for Grok 4 and 7.8% for Llama 4 Scout.
  • Llama 4 Scout has downloadable open weights; the other is API-only.

Side by side

Grok 4Llama 4 Scout
ProviderxAIMeta
Noometry Index48.127.7
Released2025-07-092025-04-05
WeightsProprietaryOpen
Context window—128K
Max output—4K
Input $ / M tokens—$0.10
Output $ / M tokens—$0.30
Results tracked4843

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

Category by category

Coding Grok 4 leads

Grok 4: 50.3 (#46), Llama 4 Scout: 20.2 (#339)

BenchmarkGrok 4Llama 4 Scout
LMArena Coding14081286
SWE-bench Verified (bash only)—9.1%
Aider Polyglot79.6%—
SciCode—17%
WeirdML45.7%—
BigCodeBench Complete—43.1%

Agentic & Tool Use Grok 4 leads

Grok 4: 32.3 (#68), Llama 4 Scout: 24.6 (#119)

BenchmarkGrok 4Llama 4 Scout
Berkeley Function Calling Leaderboard63%28.1%
Terminal-Bench27.2%—
GDPval21.1%—
Cybench43%—
DeepResearch Bench47.3%—
BALROG43.6%—
LMArena Search1142—
METR Time Horizons66.6%—

Reasoning Grok 4 leads

Grok 4: 36.7 (#65), Llama 4 Scout: 9.1 (#345)

BenchmarkGrok 4Llama 4 Scout
ARC-AGI-216%0%
Kagi LLM Benchmark73.6%36.9%
ARC-AGI-166.7%0.5%
LMArena Hard Prompts14091266
Epoch Capabilities Index146.44129.64
ForecastBench60.957.5
SimpleBench60.5%—
CritPt—0%
Chess Puzzles28%—
DTBench—57.9%
LMCA—12%

Math Grok 4 leads

Grok 4: 48.4 (#64), Llama 4 Scout: 19.6 (#286)

BenchmarkGrok 4Llama 4 Scout
OTIS Mock AIME 2024-202584%7.8%
Omni-MATH60.3%37.3%
LMArena Math14221287
FrontierMath (Feb 2025 set)19.7%0%
MATH Level 5—62.3%
FrontierMath Tier 4 (v1)2.1%—

Knowledge Grok 4 leads

Grok 4: 53.8 (#55), Llama 4 Scout: 31.9 (#217)

BenchmarkGrok 4Llama 4 Scout
GPQA Diamond87%51.8%
MMLU-Pro85.1%74.2%
GPQA (HELM)72.7%50.7%
LMArena Expert14151235
Confabulations12.4%—
Vectara Hallucination Rate—7.7%

Multimodal Grok 4 leads

Grok 4: 33.7 (#94), Llama 4 Scout: 32.2 (#102)

BenchmarkGrok 4Llama 4 Scout
LMArena Vision12101118
GeoBench45%—
SpatialViz-Bench—34.2%

Multilingual Grok 4 leads

Grok 4: 51.8 (#103), Llama 4 Scout: 41.0 (#212)

BenchmarkGrok 4Llama 4 Scout
LMArena Non-English14031252
LMArena Chinese14271255
LMArena French14181282
LMArena German14291272
LMArena Japanese13941206
LMArena Korean13771207
LMArena Russian14101263
LMArena Spanish14201278

Instruction Following Grok 4 leads

Grok 4: 79.2 (#5), Llama 4 Scout: 65.8 (#217)

BenchmarkGrok 4Llama 4 Scout
IFEval94.9%81.8%
LMArena Instruction Following13871248

Long Context Grok 4 leads

Grok 4: 63.1 (#4), Llama 4 Scout: 27.5 (#294)

BenchmarkGrok 4Llama 4 Scout
Fiction.LiveBench94.4%36%
LMArena Longer Query14091265

Writing & Preference Grok 4 leads

Grok 4: 58.5 (#116), Llama 4 Scout: 37.0 (#261)

BenchmarkGrok 4Llama 4 Scout
LMArena Text14111279
LMArena Creative Writing13971249
WildBench79.7%78%
LMArena Multi-Turn14161280
Short-Story Creative Writing76.9%—
EQ-Bench Creative Writing—783

Frequently asked questions

Is Grok 4 better than Llama 4 Scout?

Grok 4 is the stronger model overall, scoring 48.1 to 27.7 on the Noometry Index.

Is Grok 4 or Llama 4 Scout better for coding?

Grok 4 scores higher on coding benchmarks: 50.3 versus 20.2 in the Noometry coding category.

How many benchmarks do Grok 4 and Llama 4 Scout share?

33 benchmarks have published results for both models. Grok 4 has 48 scored results on Noometry and Llama 4 Scout has 43.

Related comparisons

Go deeper