← All benchmarks

Llama 3.1 Instruct 8B

Meta · Released 2024-07-23 · Independent benchmark scores from Artificial Analysis

Intelligence Index
7.6
Output speed
614.7
tokens/sec
Blended price
$0.08
per 1M tokens
Time to first token
0.22s

Model metadata

Model ID739684ba-0f63-4e2a-b4ee-30741c9e9320
Slugllama-3-1-instruct-8b
Release date2024-07-23
ProviderMeta
Provider slugmeta
Output speed614.7 tok/s
Time to first token0.22s
Time to first answer0.22s
Blended price ($/1M)$0.08
Input price ($/1M)$0.07
Output price ($/1M)$0.09

All evaluation scores (17)

aime7.7%
aime 254.3%
artificial analysis coding index5.40
artificial analysis intelligence index7.60
artificial analysis math index4.30
gpqa25.9%
hle5.1%
ifbench28.6%
lcr15.7%
livecodebench11.6%
math 50051.9%
mmlu pro47.6%
scicode13.2%
tau banking1.6%
tau216.4%
terminalbench hard0.8%
terminalbench v2 11.5%

Source: artificialanalysis.ai/models/llama-3-1-instruct-8b

Benchmark data from Artificial Analysis.