Performance Benchmarks for GPT-4 Turbo Preview (0125)

Analyze how GPT-4 Turbo Preview (0125) stacks up against other frontier models in terms of cost-to-performance and raw intelligence (LMArena Elo).

2026 Performance Summary

🚀 Intelligence (Elo): 1313
💰 API Cost (Input): $0.00001 / 1M tokens
⚡ API Cost (Output): $0.00003 / 1M tokens
⏱️ Real-world Throughput: 26 tokens/s

Analysis: While GPT-4 Turbo Preview (0125) is a capable model, it is currently outperformed in efficiency by DeepSeek-V4.1-Flash-20261017 (Max). DeepSeek-V4.1-Flash-20261017 (Max) sits higher on the Pareto line, offering approximately 351 points more intelligence (Elo) for a comparable or lower API cost.

Highly Recommended Alternatives

Frontier Model	LMArena Elo	API Cost (1M)	Throughput
DeepSeek-V4.1-Flash-20261017 (Max)	1664	$3e-7	161
Claude Haiku 5.5 (High)	1586	$1e-7	165
GLM-5.3 (Max)	1481	$7.9e-8	132
GLM-5.2 (Max)	1476	$3.41e-8	198
gpt-oss-20b	1318	$1.8e-8	30

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