Noometry

Model comparison

Grok 4.3 vs Llama 2-70B

Grok 4.3 is the stronger model overall, scoring 43.8 to 24.4 on the Noometry Index.

Last verified . 22 shared benchmarks.

Grok 4.3 xAI

43.8

Rank #86 Confirmed

Llama 2-70B Meta

24.4

Rank #349 Confirmed

Summary

  • They share 22 benchmarks with published results for both. Grok 4.3 scores higher in 8 categories and Llama 2-70B in 0 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where Grok 4.3 leads 52.5 to 7.4.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 93.3% for Grok 4.3 and 0% for Llama 2-70B.
  • Llama 2-70B has downloadable open weights; the other is API-only.

Side by side

Grok 4.3Llama 2-70B
ProviderxAIMeta
Noometry Index43.824.4
Released2026-04-172023-07-18
WeightsProprietaryOpen
Context window1M—
Max output30K—
Input $ / M tokens$1.25—
Output $ / M tokens$2.50—
Results tracked4035

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

Category by category

Coding Grok 4.3 leads

Grok 4.3: 41.6 (#121), Llama 2-70B: 31.4 (#286)

BenchmarkGrok 4.3Llama 2-70B
LMArena Coding14151079
LMArena WebDev1357—
SciCode47.3%—
WeirdML49.9%—
ALE-Bench944.17—

Agentic & Tool Use Not comparable

Grok 4.3: 27.7 (#99), Llama 2-70B: —

BenchmarkGrok 4.3Llama 2-70B
GDP.pdf8%—
LMArena Search1165—
Vending-Bench 235.26—

Reasoning Grok 4.3 leads

Grok 4.3: 35.9 (#68), Llama 2-70B: 14.4 (#325)

BenchmarkGrok 4.3Llama 2-70B
LMArena Hard Prompts13961073
DTBench90.7%41.6%
Epoch Capabilities Index149.16113.79
ForecastBench60.351.4
NYT Connections (extended)55.2%—
CritPt8%—
Chess Puzzles25%—
LMCA38.3%—
BIG-Bench Hard—64.9%
CommonsenseQA 2.0—50%
HellaSwag—85.3%
LAMBADA—78.9%
PIQA—82.8%
WinoGrande—80.2%

Math Grok 4.3 leads

Grok 4.3: 46.0 (#74), Llama 2-70B: 8.1 (#326)

BenchmarkGrok 4.3Llama 2-70B
OTIS Mock AIME 2024-202593.3%0%
LMArena Math13881091
FrontierMath (Tiers 1-3)42.8%—
FrontierMath Tier 414.6%—
ProofBench11%—
MATH Level 5—3.3%
GSM8K—69.6%

Knowledge Grok 4.3 leads

Grok 4.3: 52.5 (#62), Llama 2-70B: 7.4 (#310)

BenchmarkGrok 4.3Llama 2-70B
GPQA Diamond88.8%26.3%
LMArena Expert13851039
SimpleQA Verified33.2%—
ARC (AI2) Challenge—78.3%
BoolQ—88.6%
MMLU—69.9%
OpenBookQA—60.2%
TriviaQA—87.6%

Multimodal Not comparable

Grok 4.3: 31.6 (#104), Llama 2-70B: —

BenchmarkGrok 4.3Llama 2-70B
LMArena Vision1229—
Blueprint-Bench 20%—

Multilingual Grok 4.3 leads

Grok 4.3: 50.5 (#120), Llama 2-70B: 27.7 (#274)

BenchmarkGrok 4.3Llama 2-70B
LMArena Non-English13851045
LMArena Chinese1422995
LMArena French14121090
LMArena German13951041
LMArena Japanese1379927
LMArena Korean1356964
LMArena Russian13991083
LMArena Spanish13981143

Instruction Following Grok 4.3 leads

Grok 4.3: 72.1 (#140), Llama 2-70B: 54.9 (#278)

BenchmarkGrok 4.3Llama 2-70B
LMArena Instruction Following13661071

Long Context Grok 4.3 leads

Grok 4.3: 42.5 (#123), Llama 2-70B: 32.3 (#270)

BenchmarkGrok 4.3Llama 2-70B
LMArena Longer Query13931062

Writing & Preference Grok 4.3 leads

Grok 4.3: 58.5 (#118), Llama 2-70B: 32.3 (#279)

BenchmarkGrok 4.3Llama 2-70B
LMArena Text13971115
LMArena Creative Writing13801075
LMArena Multi-Turn14061088
EQ-Bench 41075—

Frequently asked questions

Is Grok 4.3 better than Llama 2-70B?

Grok 4.3 is the stronger model overall, scoring 43.8 to 24.4 on the Noometry Index.

Is Grok 4.3 or Llama 2-70B better for coding?

Grok 4.3 scores higher on coding benchmarks: 41.6 versus 31.4 in the Noometry coding category.

How many benchmarks do Grok 4.3 and Llama 2-70B share?

22 benchmarks have published results for both models. Grok 4.3 has 40 scored results on Noometry and Llama 2-70B has 35.

Related comparisons

Go deeper