Noometry

Model comparison

Grok 4.20 Multi-Agent vs Llama 2-70B

Grok 4.20 Multi-Agent is the stronger model overall, scoring 46.2 to 24.4 on the Noometry Index.

Last verified . 17 shared benchmarks.

Grok 4.20 Multi-Agent xAI

46.2

Rank #65 Confirmed

Llama 2-70B Meta

24.4

Rank #349 Confirmed

Summary

  • They share 17 benchmarks with published results for both. Grok 4.20 Multi-Agent 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 4.20 Multi-Agent leads 40.4 to 7.4.
  • Llama 2-70B has downloadable open weights; the other is API-only.

Side by side

Grok 4.20 Multi-AgentLlama 2-70B
ProviderxAIMeta
Noometry Index46.224.4
Released2026-03-092023-07-18
WeightsProprietaryOpen
Context window1M—
Max output30K—
Input $ / M tokens$1.25—
Output $ / M tokens$2.50—
Results tracked2035

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

Category by category

Coding Grok 4.20 Multi-Agent leads

Grok 4.20 Multi-Agent: 43.0 (#92), Llama 2-70B: 31.4 (#286)

BenchmarkGrok 4.20 Multi-AgentLlama 2-70B
LMArena Coding14571079

Agentic & Tool Use Not comparable

Grok 4.20 Multi-Agent: —, Llama 2-70B: —

BenchmarkGrok 4.20 Multi-AgentLlama 2-70B
LMArena Search1204—

Reasoning Grok 4.20 Multi-Agent leads

Grok 4.20 Multi-Agent: 43.9 (#48), Llama 2-70B: 14.4 (#325)

BenchmarkGrok 4.20 Multi-AgentLlama 2-70B
LMArena Hard Prompts14481073
NYT Connections (extended)89.6%—
DTBench—41.6%
BIG-Bench Hard—64.9%
CommonsenseQA 2.0—50%
Epoch Capabilities Index—113.79
ForecastBench—51.4
HellaSwag—85.3%
LAMBADA—78.9%
PIQA—82.8%
WinoGrande—80.2%

Math Grok 4.20 Multi-Agent leads

Grok 4.20 Multi-Agent: 39.4 (#104), Llama 2-70B: 8.1 (#326)

BenchmarkGrok 4.20 Multi-AgentLlama 2-70B
LMArena Math14421091
OTIS Mock AIME 2024-2025—0%
MATH Level 5—3.3%
GSM8K—69.6%

Knowledge Grok 4.20 Multi-Agent leads

Grok 4.20 Multi-Agent: 40.4 (#119), Llama 2-70B: 7.4 (#310)

BenchmarkGrok 4.20 Multi-AgentLlama 2-70B
LMArena Expert14451039
GPQA Diamond—26.3%
ARC (AI2) Challenge—78.3%
BoolQ—88.6%
MMLU—69.9%
OpenBookQA—60.2%
TriviaQA—87.6%

Multimodal Not comparable

Grok 4.20 Multi-Agent: 40.5 (#48), Llama 2-70B: —

BenchmarkGrok 4.20 Multi-AgentLlama 2-70B
LMArena Vision1259—

Multilingual Grok 4.20 Multi-Agent leads

Grok 4.20 Multi-Agent: 54.4 (#43), Llama 2-70B: 27.7 (#274)

BenchmarkGrok 4.20 Multi-AgentLlama 2-70B
LMArena Non-English14401045
LMArena Chinese1475995
LMArena French14661090
LMArena German14561041
LMArena Japanese1405927
LMArena Korean1416964
LMArena Russian14571083
LMArena Spanish14471143

Instruction Following Grok 4.20 Multi-Agent leads

Grok 4.20 Multi-Agent: 74.8 (#84), Llama 2-70B: 54.9 (#278)

BenchmarkGrok 4.20 Multi-AgentLlama 2-70B
LMArena Instruction Following14201071

Long Context Grok 4.20 Multi-Agent leads

Grok 4.20 Multi-Agent: 43.7 (#88), Llama 2-70B: 32.3 (#270)

BenchmarkGrok 4.20 Multi-AgentLlama 2-70B
LMArena Longer Query14311062

Writing & Preference Grok 4.20 Multi-Agent leads

Grok 4.20 Multi-Agent: 64.0 (#59), Llama 2-70B: 32.3 (#279)

BenchmarkGrok 4.20 Multi-AgentLlama 2-70B
LMArena Text14501115
LMArena Creative Writing14361075
LMArena Multi-Turn14521088

Frequently asked questions

Is Grok 4.20 Multi-Agent better than Llama 2-70B?

Grok 4.20 Multi-Agent is the stronger model overall, scoring 46.2 to 24.4 on the Noometry Index.

Is Grok 4.20 Multi-Agent or Llama 2-70B better for coding?

Grok 4.20 Multi-Agent scores higher on coding benchmarks: 43.0 versus 31.4 in the Noometry coding category.

How many benchmarks do Grok 4.20 Multi-Agent and Llama 2-70B share?

17 benchmarks have published results for both models. Grok 4.20 Multi-Agent has 20 scored results on Noometry and Llama 2-70B has 35.

Related comparisons

Go deeper