Performance Benchmarks for granite-4.2-8b

Analyze how granite-4.2-8b stacks up against other frontier models in terms of cost-to-performance and raw intelligence (LMArena Elo).

2026 Performance Summary

🚀 Intelligence (Elo): 1288
💰 API Cost (Input): $1e-7 / 1M tokens
⚡ API Cost (Output): $1.5e-7 / 1M tokens
⏱️ Real-world Throughput: 85 tokens/s

Analysis: While granite-4.2-8b is a capable model, it is currently outperformed in efficiency by hy4-preview. hy4-preview sits higher on the Pareto line, offering approximately 338 points more intelligence (Elo) for a comparable or lower API cost.

Highly Recommended Alternatives

Frontier Model	LMArena Elo	API Cost (1M)	Throughput
hy4-preview	1626	$8.34e-7	38
glm-5.3-flash	1474	$7.5e-8	90
gemma-4-26b-a4b	1438	$7e-8	43
qwen3.5-flash	1397	$6.5e-8	39
qwen3-30b-a3b-instruct-2507	1382	$4.815e-8	60

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