← All comparisons

Llama 3.1 Instruct 70B vs Qwen3 VL 235B A22B (Reasoning)

Meta vs Alibaba — side-by-side benchmark comparison

Llama 3.1 Instruct 70BQwen3 VL 235B A22B (Reasoning)
Intelligence Index12.527.6
Coding Index10.920.9
Math Index4.088.3
Output speed (tok/s)35.335.6
Blended price ($/1M)$0.56$2.17
Time to first token (s)0.54s5.14s
aime17.3%
aime 254.0%88.3%
artificial analysis coding index10.9020.90
artificial analysis intelligence index12.5027.60
artificial analysis math index4.0088.30
gpqa40.9%77.2%
hle4.6%10.1%
ifbench34.4%56.5%
lcr6.3%58.7%
livecodebench23.2%64.6%
math 50064.9%
mmlu pro67.6%83.6%
scicode26.7%39.9%
tau215.2%54.1%
terminalbench hard3.0%11.4%

Benchmark data from Artificial Analysis.