← All comparisons

Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning) vs Qwen3 VL 32B (Reasoning)

Google vs Alibaba — side-by-side benchmark comparison

Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)Qwen3 VL 32B (Reasoning)
Intelligence Index21.624.7
Coding Index18.214.5
Math Index68.784.7
Output speed (tok/s)0.096.3
Blended price ($/1M)$0.17$2.63
Time to first token (s)0.00s1.12s
aime
aime 2568.7%84.7%
artificial analysis coding index18.2014.50
artificial analysis intelligence index21.6024.70
artificial analysis math index68.7084.70
gpqa70.9%73.3%
hle6.6%9.6%
ifbench52.6%59.4%
lcr59.0%55.3%
livecodebench68.8%73.8%
math 500
mmlu pro80.8%81.8%
scicode28.7%28.5%
tau230.7%45.6%
terminalbench hard12.9%7.6%

Benchmark data from Artificial Analysis.