Performance Benchmarks for deepseek-v4-pro

Analyze how deepseek-v4-pro stacks up against other frontier models in terms of cost-to-performance and raw intelligence (LMArena Elo).

2026 Performance Summary

🚀 Intelligence (Elo): 1463
💰 API Cost (Input): $0.00000174 / 1M tokens
⚡ API Cost (Output): $0.00000348 / 1M tokens
⏱️ Real-world Throughput: 25 tokens/s

Analysis: While deepseek-v4-pro is a capable model, it is currently outperformed in efficiency by claude-opus-4-7-thinking. claude-opus-4-7-thinking sits higher on the Pareto line, offering approximately 40 points more intelligence (Elo) for a comparable or lower API cost.

Highly Recommended Alternatives

Frontier Model	LMArena Elo	API Cost (1M)	Throughput
claude-opus-4-7-thinking	1503	$0.000005	52
gemini-3.1-pro-preview	1493	$0.000002	85
gpt-5.2-chat-latest-20260210	1476	$0.00000175	25
gemini-3-flash	1474	$5e-7	66
gemma-4-31b	1451	$1.3e-7	19

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