← All comparisons

Llama 3 Instruct 8B vs Claude 3.5 Sonnet (June '24)

Meta vs Anthropic — side-by-side benchmark comparison

Llama 3 Instruct 8BClaude 3.5 Sonnet (June '24)
Intelligence Index6.414.2
Coding Index4.026.0
Math Index
Output speed (tok/s)83.30.0
Blended price ($/1M)$0.07$6.56
Time to first token (s)0.47s0.00s
aime0.0%9.7%
aime 25
artificial analysis coding index4.0026.00
artificial analysis intelligence index6.4014.20
artificial analysis math index
gpqa29.6%56.0%
hle5.1%3.7%
ifbench24.6%
lcr0.0%
livecodebench9.6%
math 50049.9%69.5%
mmlu pro40.5%75.1%
scicode11.9%31.6%
tau20.0%
terminalbench hard0.0%

Benchmark data from Artificial Analysis.