Noometry

Model comparison

Grok 4 Fast vs Llama 2-70B

Grok 4 Fast is the stronger model overall, scoring 39.4 to 24.4 on the Noometry Index.

Last verified . 20 shared benchmarks.

Grok 4 Fast xAI

39.4

Rank #167 Confirmed

Llama 2-70B Meta

24.4

Rank #349 Confirmed

Summary

  • They share 20 benchmarks with published results for both. Grok 4 Fast scores higher in 8 categories and Llama 2-70B in 0 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in long context, where Grok 4 Fast leads 63.2 to 32.3.
  • The biggest single-benchmark swing is DTBench: 82.7% for Grok 4 Fast and 41.6% for Llama 2-70B.
  • Llama 2-70B has downloadable open weights; the other is API-only.

Side by side

Grok 4 FastLlama 2-70B
ProviderxAIMeta
Noometry Index39.424.4
Released2025-09-192023-07-18
WeightsProprietaryOpen
Context window——
Max output——
Input $ / M tokens——
Output $ / M tokens——
Results tracked3035

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

Category by category

Coding Grok 4 Fast leads

Grok 4 Fast: 32.5 (#271), Llama 2-70B: 31.4 (#286)

BenchmarkGrok 4 FastLlama 2-70B
LMArena Coding14291079
LMArena WebDev1159—
WeirdML42.9%—

Agentic & Tool Use Not comparable

Grok 4 Fast: 29.5 (#86), Llama 2-70B: —

BenchmarkGrok 4 FastLlama 2-70B
τ²-bench Banking15.7%—
Cybench30%—
LMArena Search1171—

Reasoning Grok 4 Fast leads

Grok 4 Fast: 22.2 (#201), Llama 2-70B: 14.4 (#325)

BenchmarkGrok 4 FastLlama 2-70B
LMArena Hard Prompts14121073
DTBench82.7%41.6%
Epoch Capabilities Index144.2113.79
ForecastBench60.551.4
ARC-AGI-25.3%—
Kagi LLM Benchmark66.1%—
ARC-AGI-148.5%—
BIG-Bench Hard—64.9%
CommonsenseQA 2.0—50%
HellaSwag—85.3%
LAMBADA—78.9%
PIQA—82.8%
WinoGrande—80.2%

Math Grok 4 Fast leads

Grok 4 Fast: 38.9 (#123), Llama 2-70B: 8.1 (#326)

BenchmarkGrok 4 FastLlama 2-70B
LMArena Math14191091
OTIS Mock AIME 2024-2025—0%
MATH Level 5—3.3%
GSM8K—69.6%

Knowledge Grok 4 Fast leads

Grok 4 Fast: 32.0 (#214), Llama 2-70B: 7.4 (#310)

BenchmarkGrok 4 FastLlama 2-70B
LMArena Expert14111039
GPQA Diamond—26.3%
Vectara Hallucination Rate19.7%—
ARC (AI2) Challenge—78.3%
BoolQ—88.6%
MMLU—69.9%
OpenBookQA—60.2%
TriviaQA—87.6%

Multilingual Grok 4 Fast leads

Grok 4 Fast: 51.3 (#111), Llama 2-70B: 27.7 (#274)

BenchmarkGrok 4 FastLlama 2-70B
LMArena Non-English13961045
LMArena Chinese1457995
LMArena French14311090
LMArena German13831041
LMArena Japanese1352927
LMArena Korean1357964
LMArena Russian13891083
LMArena Spanish14151143

Instruction Following Grok 4 Fast leads

Grok 4 Fast: 73.2 (#121), Llama 2-70B: 54.9 (#278)

BenchmarkGrok 4 FastLlama 2-70B
LMArena Instruction Following13871071

Long Context Grok 4 Fast leads

Grok 4 Fast: 63.2 (#3), Llama 2-70B: 32.3 (#270)

BenchmarkGrok 4 FastLlama 2-70B
LMArena Longer Query14151062
Fiction.LiveBench94.4%—

Writing & Preference Grok 4 Fast leads

Grok 4 Fast: 60.0 (#102), Llama 2-70B: 32.3 (#279)

BenchmarkGrok 4 FastLlama 2-70B
LMArena Text14071115
LMArena Creative Writing13871075
LMArena Multi-Turn14141088

Frequently asked questions

Is Grok 4 Fast better than Llama 2-70B?

Grok 4 Fast is the stronger model overall, scoring 39.4 to 24.4 on the Noometry Index.

Is Grok 4 Fast or Llama 2-70B better for coding?

Grok 4 Fast scores higher on coding benchmarks: 32.5 versus 31.4 in the Noometry coding category.

How many benchmarks do Grok 4 Fast and Llama 2-70B share?

20 benchmarks have published results for both models. Grok 4 Fast has 30 scored results on Noometry and Llama 2-70B has 35.

Related comparisons

Go deeper