← All comparisons

Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) vs Qwen Chat 72B

Anthropic vs Alibaba — side-by-side benchmark comparison

Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)Qwen Chat 72B
Intelligence Index51.78.8
Coding Index50.9
Math Index
Output speed (tok/s)68.20.0
Blended price ($/1M)$6.56$0.00
Time to first token (s)55.35s0.00s
aime
aime 25
artificial analysis coding index50.90
artificial analysis intelligence index51.708.80
artificial analysis math index
gpqa87.5%
hle30.0%
ifbench56.6%
lcr70.7%
livecodebench
math 500
mmlu pro
scicode46.8%
tau275.7%
terminalbench hard53.0%

Benchmark data from Artificial Analysis.