modelbenchmark.io

llama-2-70b

Meta · llama-2-70b

Compare

Specification

most-agreed values

Context
4K
Max output
4K
Released
Knowledge cutoff
Retires
Open weights
Input
Output

Price

US dollars per million tokens · most-agreed

Input
$0.65
Output
$2.75

Quality

3 benchmarks · 1 source

Composite
2nd
percentile of 334 scored models
Rank
326 / 334
±0.06 sd
Evidence
3 × 1
one source, or fewer than 3 benchmarks
Effort range
one configuration only
reasoning1st1/4 bench
math0th2/7 bench

Every score

one row per source and configuration — nothing averaged away

BenchmarkScoreConfigurationSourceRun
OTIS Mock AIME 2024-20250.0 ±0.0defaultEpoch AI2025-02-25
MATH level 53.3 ±0.3defaultEpoch AI2025-01-27
GPQA diamond26.3 ±2.0defaultEpoch AI2025-01-27

The leading figure for a benchmark is the median across its configurations, so one heroic high-effort run cannot set the number.

Available from 1 host

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
replicate · replicate$0.65$2.754K4K

More from Meta

most-hosted first