Noometry

Model comparison

Grok 4.6 vs Llama 2-70B

Grok 4.6 is the stronger model overall, scoring 56.9 to 24.4 on the Noometry Index.

Last verified . 21 shared benchmarks.

Grok 4.6 xAI

56.9

Rank #21 Confirmed

Llama 2-70B Meta

24.4

Rank #349 Confirmed

Summary

  • They share 21 benchmarks with published results for both. Grok 4.6 scores higher in 8 categories and Llama 2-70B in 0 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where Grok 4.6 leads 67.0 to 8.1.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 99.2% for Grok 4.6 and 0% for Llama 2-70B.
  • Llama 2-70B has downloadable open weights; the other is API-only.

Side by side

Grok 4.6Llama 2-70B
ProviderxAIMeta
Noometry Index56.924.4
Released2026-08-122023-07-18
WeightsProprietaryOpen
Context window500K—
Max output500K—
Input $ / M tokens$2—
Output $ / M tokens$6—
Results tracked4935

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

Category by category

Coding Grok 4.6 leads

Grok 4.6: 58.5 (#16), Llama 2-70B: 31.4 (#286)

BenchmarkGrok 4.6Llama 2-70B
LMArena Coding14651079
DeepSWE67.5%—
FrontierCode48%—
CursorBench41.4%—
LMArena WebDev1617—
FrontierSWE25.3%—
SciCode56.5%—
WeirdML67.3%—
ALE-Bench1,508—

Agentic & Tool Use Not comparable

Grok 4.6: 39.4 (#27), Llama 2-70B: —

BenchmarkGrok 4.6Llama 2-70B
APEX-Agents65.3%—
GDP.pdf17.2%—
Vending-Bench 29,047—

Reasoning Grok 4.6 leads

Grok 4.6: 61.4 (#20), Llama 2-70B: 14.4 (#325)

BenchmarkGrok 4.6Llama 2-70B
LMArena Hard Prompts14471073
DTBench97.3%41.6%
Epoch Capabilities Index156.44113.79
ARC-AGI-267.1%—
SimpleBench75.9%—
NYT Connections (extended)80%—
ARC-AGI-187.5%—
CritPt19.7%—
Chess Puzzles40%—
EBR-Bench30.5%—
Mystery Game Puzzles34%—
LMCA48.5%—
BIG-Bench Hard—64.9%
CommonsenseQA 2.0—50%
ForecastBench—51.4
HellaSwag—85.3%
LAMBADA—78.9%
PIQA—82.8%
WinoGrande—80.2%

Math Grok 4.6 leads

Grok 4.6: 67.0 (#24), Llama 2-70B: 8.1 (#326)

BenchmarkGrok 4.6Llama 2-70B
OTIS Mock AIME 2024-202599.2%0%
LMArena Math14231091
FrontierMath (Tiers 1-3)66%—
FrontierMath Tier 431.7%—
ProofBench51%—
MATH Level 5—3.3%
GSM8K—69.6%

Knowledge Grok 4.6 leads

Grok 4.6: 63.3 (#20), Llama 2-70B: 7.4 (#310)

BenchmarkGrok 4.6Llama 2-70B
GPQA Diamond94%26.3%
LMArena Expert14671039
SimpleQA Verified49.3%—
ARC (AI2) Challenge—78.3%
BoolQ—88.6%
MMLU—69.9%
OpenBookQA—60.2%
TriviaQA—87.6%

Multimodal Not comparable

Grok 4.6: 43.6 (#23), Llama 2-70B: —

BenchmarkGrok 4.6Llama 2-70B
LMArena Vision1263—
Blueprint-Bench 233.2%—
Furniture Assembly40%—
LMArena Document1452—

Multilingual Grok 4.6 leads

Grok 4.6: 53.0 (#74), Llama 2-70B: 27.7 (#274)

BenchmarkGrok 4.6Llama 2-70B
LMArena Non-English14201045
LMArena Chinese1480995
LMArena French14611090
LMArena German14311041
LMArena Japanese1376927
LMArena Korean1397964
LMArena Russian14221083
LMArena Spanish14041143

Instruction Following Grok 4.6 leads

Grok 4.6: 75.4 (#63), Llama 2-70B: 54.9 (#278)

BenchmarkGrok 4.6Llama 2-70B
LMArena Instruction Following14311071

Long Context Grok 4.6 leads

Grok 4.6: 44.5 (#66), Llama 2-70B: 32.3 (#270)

BenchmarkGrok 4.6Llama 2-70B
LMArena Longer Query14541062

Writing & Preference Grok 4.6 leads

Grok 4.6: 62.3 (#80), Llama 2-70B: 32.3 (#279)

BenchmarkGrok 4.6Llama 2-70B
LMArena Text14281115
LMArena Creative Writing14281075
LMArena Multi-Turn14251088

Frequently asked questions

Is Grok 4.6 better than Llama 2-70B?

Grok 4.6 is the stronger model overall, scoring 56.9 to 24.4 on the Noometry Index.

Is Grok 4.6 or Llama 2-70B better for coding?

Grok 4.6 scores higher on coding benchmarks: 58.5 versus 31.4 in the Noometry coding category.

How many benchmarks do Grok 4.6 and Llama 2-70B share?

21 benchmarks have published results for both models. Grok 4.6 has 49 scored results on Noometry and Llama 2-70B has 35.

Related comparisons

Go deeper