Noometry

Model comparison

Grok 3 vs Llama 3.1-70B

Grok 3 is the stronger model overall, scoring 39.9 to 29.6 on the Noometry Index.

Last verified . 29 shared benchmarks.

Grok 3 xAI

39.9

Rank #157 Confirmed

Llama 3.1-70B Meta

29.6

Rank #308 Confirmed

Summary

  • They share 29 benchmarks with published results for both. Grok 3 scores higher in 8 categories and Llama 3.1-70B in 1 category; 9 gaps are clear of the uncertainty.
  • The widest gap is in math, where Grok 3 leads 38.0 to 13.5.
  • The biggest single-benchmark swing is MATH Level 5: 88.7% for Grok 3 and 36.7% for Llama 3.1-70B.
  • Llama 3.1-70B has downloadable open weights; the other is API-only.

Side by side

Grok 3Llama 3.1-70B
ProviderxAIMeta
Noometry Index39.929.6
Released2025-04-092024-07-23
WeightsProprietaryOpen
Context window—128K
Max output—4K
Input $ / M tokens—$0.40
Output $ / M tokens—$0.40
Results tracked4035

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

Category by category

Coding Grok 3 leads

Grok 3: 41.9 (#115), Llama 3.1-70B: 30.3 (#296)

BenchmarkGrok 3Llama 3.1-70B
WeirdML37.2%9%
LMArena Coding14321260
Aider Polyglot53.3%—
BigCodeBench Instruct—46.1%
BigCodeBench Complete—54.8%

Agentic & Tool Use Grok 3 leads

Grok 3: 30.5 (#76), Llama 3.1-70B: 25.1 (#112)

BenchmarkGrok 3Llama 3.1-70B
BALROG29.5%27.9%
TheAgentCompany—6.9%

Reasoning Llama 3.1-70B leads

Grok 3: 13.7 (#333), Llama 3.1-70B: 21.6 (#220)

BenchmarkGrok 3Llama 3.1-70B
LMArena Hard Prompts14341241
Epoch Capabilities Index138.33125.92
ARC-AGI-20%—
SimpleBench36.1%—
Kagi LLM Benchmark61.3%—
ARC-AGI-15.5%—
DTBench—60%
LMCA—14.8%

Math Grok 3 leads

Grok 3: 38.0 (#145), Llama 3.1-70B: 13.5 (#304)

BenchmarkGrok 3Llama 3.1-70B
OTIS Mock AIME 2024-202555.6%3.6%
Omni-MATH46.4%21%
LMArena Math13911252
MATH Level 588.7%36.7%
FrontierMath (Feb 2025 set)3.8%—
FrontierMath Tier 4 (v1)0%—

Knowledge Grok 3 leads

Grok 3: 46.2 (#82), Llama 3.1-70B: 24.2 (#269)

BenchmarkGrok 3Llama 3.1-70B
GPQA Diamond75.8%44.2%
MMLU-Pro78.8%65.3%
GPQA (HELM)65%42.6%
LMArena Expert14211209
Confabulations14.2%—
Vectara Hallucination Rate5.8%—
MMLU—80.1%

Multilingual Grok 3 leads

Grok 3: 52.3 (#87), Llama 3.1-70B: 38.8 (#225)

BenchmarkGrok 3Llama 3.1-70B
LMArena Non-English14101219
LMArena Chinese14481215
LMArena French14601261
LMArena German14311222
LMArena Japanese13871132
LMArena Korean13731140
LMArena Russian14161234
LMArena Spanish14171253

Instruction Following Grok 3 leads

Grok 3: 75.0 (#73), Llama 3.1-70B: 65.3 (#223)

BenchmarkGrok 3Llama 3.1-70B
IFEval88.4%82.1%
LMArena Instruction Following14091231

Long Context Grok 3 leads

Grok 3: 38.7 (#192), Llama 3.1-70B: 37.6 (#214)

BenchmarkGrok 3Llama 3.1-70B
LMArena Longer Query14391241
Fiction.LiveBench58.3%—

Writing & Preference Grok 3 leads

Grok 3: 55.8 (#141), Llama 3.1-70B: 35.4 (#267)

BenchmarkGrok 3Llama 3.1-70B
LMArena Text14261261
LMArena Creative Writing14141232
EQ-Bench Creative Writing1186784
WildBench84.9%75.8%
LMArena Multi-Turn14251256
Short-Story Creative Writing76.4%—

Frequently asked questions

Is Grok 3 better than Llama 3.1-70B?

Grok 3 is the stronger model overall, scoring 39.9 to 29.6 on the Noometry Index.

Is Grok 3 or Llama 3.1-70B better for coding?

Grok 3 scores higher on coding benchmarks: 41.9 versus 30.3 in the Noometry coding category.

How many benchmarks do Grok 3 and Llama 3.1-70B share?

29 benchmarks have published results for both models. Grok 3 has 40 scored results on Noometry and Llama 3.1-70B has 35.

Related comparisons

Go deeper