← All comparisons

Grok 4.20 0309 (Reasoning) vs Qwen3 VL 235B A22B (Reasoning)

xAI vs Alibaba — side-by-side benchmark comparison

Grok 4.20 0309 (Reasoning)Qwen3 VL 235B A22B (Reasoning)
Intelligence Index48.527.6
Coding Index42.220.9
Math Index88.3
Output speed (tok/s)217.835.6
Blended price ($/1M)$3.00$2.17
Time to first token (s)13.18s5.14s
aime
aime 2588.3%
artificial analysis coding index42.2020.90
artificial analysis intelligence index48.5027.60
artificial analysis math index88.30
gpqa88.5%77.2%
hle30.0%10.1%
ifbench82.9%56.5%
lcr59.0%58.7%
livecodebench64.6%
math 500
mmlu pro83.6%
scicode44.7%39.9%
tau296.5%54.1%
terminalbench hard40.9%11.4%

Benchmark data from Artificial Analysis.