Performance Benchmarks for GPT-3.5 Turbo (0125)

Analyze how GPT-3.5 Turbo (0125) stacks up against other frontier models in terms of cost-to-performance and raw intelligence (LMArena Elo).

2026 Performance Summary

🚀 Intelligence (Elo): 1226
💰 API Cost (Input): $0.000003 / 1M tokens
⚡ API Cost (Output): $0.000004 / 1M tokens
⏱️ Real-world Throughput: 79 tokens/s

Analysis: While GPT-3.5 Turbo (0125) is a capable model, it is currently outperformed in efficiency by Kimi K3.1 Pro (Max). Kimi K3.1 Pro (Max) sits higher on the Pareto line, offering approximately 466 points more intelligence (Elo) for a comparable or lower API cost.

Highly Recommended Alternatives

Frontier Model	LMArena Elo	API Cost (1M)	Throughput
Kimi K3.1 Pro (Max)	1692	$0.000002	32
DeepSeek-V4.1-Flash-20261017 (Max)	1665	$1.5e-7	190
GLM-5.3 (Max)	1481	$7.9e-8	119
DeepSeek-V4.1-Flash-20260910 (Max)	1476	$2e-8	157
gpt-oss-20b	1317	$1.8e-8	47

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