← All comparisons

Llama 3.1 Instruct 405B vs Qwen3.5 27B (Non-reasoning)

Meta vs Alibaba — side-by-side benchmark comparison

Llama 3.1 Instruct 405BQwen3.5 27B (Non-reasoning)
Intelligence Index17.437.2
Coding Index14.533.4
Math Index3.0
Output speed (tok/s)37.595.3
Blended price ($/1M)$3.69$0.88
Time to first token (s)0.63s1.40s
aime21.3%
aime 253.0%
artificial analysis coding index14.5033.40
artificial analysis intelligence index17.4037.20
artificial analysis math index3.00
gpqa51.5%84.2%
hle4.2%13.2%
ifbench39.0%46.9%
lcr24.3%55.7%
livecodebench30.5%
math 50070.3%
mmlu pro73.2%
scicode29.9%36.7%
tau219.0%87.1%
terminalbench hard6.8%31.8%

Benchmark data from Artificial Analysis.