deepseek-v3.2-exp-thinking vs gemini-2.5-pro

Direct benchmark comparison between deepseek-v3.2-exp-thinking and gemini-2.5-pro based on LMArena Elo and latest API pricing.

Direct Technical Comparison

Frontier Model	LMArena Elo	API Cost (1M)	Throughput
deepseek-v3.2-exp-thinking	1425	$2.7e-7	7
gemini-2.5-pro	1448	$0.00000125	88

*These models represent the Pareto Frontier (optimal cost-to-performance).*

Conclusion: For developers looking for higher intelligence, gemini-2.5-pro currently leads with an Elo of 1448.