← All comparisons

Claude 3.5 Haiku vs Qwen3 VL 30B A3B Instruct

Anthropic vs Alibaba — side-by-side benchmark comparison

Claude 3.5 HaikuQwen3 VL 30B A3B Instruct
Intelligence Index18.716.0
Coding Index10.714.3
Math Index72.3
Output speed (tok/s)0.0123.5
Blended price ($/1M)$1.75$0.30
Time to first token (s)0.00s1.07s
aime3.3%
aime 2572.3%
artificial analysis coding index10.7014.30
artificial analysis intelligence index18.7016.00
artificial analysis math index72.30
gpqa40.8%69.5%
hle3.5%6.4%
ifbench42.8%33.1%
lcr23.3%23.7%
livecodebench31.4%47.6%
math 50072.1%
mmlu pro63.4%76.4%
scicode27.4%30.8%
tau224.6%19.0%
terminalbench hard2.3%6.1%

Benchmark data from Artificial Analysis.