← All comparisons

Claude Sonnet 4.6 (Non-reasoning, High Effort) vs DeepSeek R1 Distill Qwen 1.5B

Anthropic vs DeepSeek — side-by-side benchmark comparison

Claude Sonnet 4.6 (Non-reasoning, High Effort)DeepSeek R1 Distill Qwen 1.5B
Intelligence Index44.49.1
Coding Index46.4
Math Index22.0
Output speed (tok/s)55.20.0
Blended price ($/1M)$6.56$0.00
Time to first token (s)1.07s0.00s
aime17.7%
aime 2522.0%
artificial analysis coding index46.40
artificial analysis intelligence index44.409.10
artificial analysis math index22.00
gpqa79.9%9.8%
hle13.2%3.3%
ifbench41.2%13.2%
lcr57.7%0.3%
livecodebench7.0%
math 50068.7%
mmlu pro26.9%
scicode46.9%6.6%
tau279.5%
terminalbench hard46.2%

Benchmark data from Artificial Analysis.