deepseek-v3.2-exp-thinking vs gemini-3.1-flash-lite-preview

Direct benchmark comparison between deepseek-v3.2-exp-thinking and gemini-3.1-flash-lite-preview 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-3.1-flash-lite-preview	1438	$2.5e-7	58

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

Conclusion: For developers looking for higher intelligence, gemini-3.1-flash-lite-preview currently leads with an Elo of 1438.