← All comparisons

GPT-5.4 mini (Non-Reasoning) vs DeepSeek R1 Distill Qwen 14B

OpenAI vs DeepSeek — side-by-side benchmark comparison

GPT-5.4 mini (Non-Reasoning)DeepSeek R1 Distill Qwen 14B
Intelligence Index23.315.8
Coding Index25.3
Math Index55.7
Output speed (tok/s)166.40.0
Blended price ($/1M)$1.69$0.00
Time to first token (s)0.48s0.00s
aime66.7%
aime 2555.7%
artificial analysis coding index25.30
artificial analysis intelligence index23.3015.80
artificial analysis math index55.70
gpqa60.6%48.4%
hle5.7%4.4%
ifbench38.8%22.1%
lcr30.3%7.0%
livecodebench37.6%
math 50094.9%
mmlu pro74.0%
scicode39.6%23.9%
tau223.4%
terminalbench hard18.2%

Benchmark data from Artificial Analysis.