Llama 2-70B

Released2023-07-18
Price per 1MNot listed
Overall score22.5 (early, few results)
ProviderMeta
WeightsOpen weights

Scores

DimensionScoreRankBenchmarks
Overall: Balanced 22.5 4
Overall: Coding 27.6 1
Overall: Reasoning 24.9 3
Overall: Agentic work 27.6 1
Reasoning 27.6 1
Math 21.4 2
General 16.0 1

Evidence

Reasoning 27.6 ±4.6

BenchmarkResultSourceDate
GPQA Diamond 26.3% Epoch AI 2025-01-27

Math 21.4 ±2.0

BenchmarkResultSourceDate
MATH Level 5 3.3% Epoch AI 2025-01-27
OTIS Mock AIME 2024-2025 0.0% Epoch AI 2025-02-25

General 16.0 ±4.1

BenchmarkResultSourceDate
LMArena Text 1,171 LMArena 2026-11-09

Names matched to this model

hfendpoint-llama-2-70b, llama-2-70b, llama-2-70b-chat, llama-2-70b-chat-hf, llama-2-70b-hf

Compare Llama 2-70B with other models