← All comparisons

gpt-oss-20B (low) vs Qwen3.6 27B (Non-reasoning)

OpenAI vs Alibaba — side-by-side benchmark comparison

gpt-oss-20B (low)Qwen3.6 27B (Non-reasoning)
Intelligence Index20.837.1
Coding Index14.426.6
Math Index62.3
Output speed (tok/s)273.060.7
Blended price ($/1M)$0.10$1.35
Time to first token (s)0.50s1.55s
aime
aime 2562.3%
artificial analysis coding index14.4026.60
artificial analysis intelligence index20.8037.10
artificial analysis math index62.30
gpqa61.1%82.9%
hle5.1%13.6%
ifbench57.8%45.7%
lcr31.0%55.0%
livecodebench65.2%
math 500
mmlu pro71.8%
scicode34.0%37.3%
tau250.3%93.6%
terminalbench hard4.5%21.2%

Benchmark data from Artificial Analysis.