Benchmarks for gemma-4-26b-a4b

Analyze how gemma-4-26b-a4b stacks up against other models in terms of cost-to-performance and raw intelligence (LMArena Elo).

Performance Summary

🚀 Intelligence (Elo): 1441
💰 Price (Input): $1.3e-7 / 1M tokens
⚡ Price (Output): $4e-7 / 1M tokens
⏱️ Throughput: 12 tokens/s

Recommended Alternatives

Frontier Model	LMArena Elo	API Cost (1M)	Throughput
gemma-4-31b	1452	$1.4e-7	2
qwen3-235b-a22b-instruct-2507	1422	$7.1e-8	6
qwen3.5-flash	1402	$6.5e-8	102
glm-4.7-flash	1369	$6e-8	50
gpt-oss-120b	1354	$3.9e-8	51

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