← All comparisons

Claude 3 Haiku vs Qwen3 30B A3B 2507 (Reasoning)

Anthropic vs Alibaba — side-by-side benchmark comparison

Claude 3 HaikuQwen3 30B A3B 2507 (Reasoning)
Intelligence Index12.322.4
Coding Index6.714.6
Math Index56.3
Output speed (tok/s)0.0155.3
Blended price ($/1M)$0.50$0.67
Time to first token (s)0.00s1.02s
aime1.0%90.7%
aime 2556.3%
artificial analysis coding index6.7014.60
artificial analysis intelligence index12.3022.40
artificial analysis math index56.30
gpqa37.4%70.7%
hle3.9%9.8%
ifbench36.1%50.7%
lcr21.0%59.0%
livecodebench15.4%70.7%
math 50039.4%97.6%
mmlu pro80.5%
scicode18.6%33.3%
tau221.1%28.1%
terminalbench hard0.8%5.3%

Benchmark data from Artificial Analysis.