Performance Benchmarks for hy4-preview

Analyze how hy4-preview stacks up against other frontier models in terms of cost-to-performance and raw intelligence (LMArena Elo).

2026 Performance Summary

🚀 Intelligence (Elo): 1631
💰 API Cost (Input): $8.34e-7 / 1M tokens
⚡ API Cost (Output): $0.000002501 / 1M tokens
⏱️ Real-world Throughput: 30 tokens/s

Analysis: hy4-preview is currently positioned on the Pareto Frontier. This means it offers a unique and optimal balance between intelligence (Elo: 1631) and cost ($8.34e-7/1M tokens), making it a top-tier choice for its price bracket.

Highly Recommended Alternatives

Frontier Model	LMArena Elo	API Cost (1M)	Throughput
gemini-3-flash	1474	$5e-7	75
glm-5.3-flash	1469	$7.5e-8	109
gemma-4-26b-a4b	1438	$7e-8	52
qwen3.5-flash	1397	$6.5e-8	41
qwen3-30b-a3b-instruct-2507	1382	$4.815e-8	79

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