Performance Benchmarks for grok-4.20-beta-0309-reasoning

Analyze how grok-4.20-beta-0309-reasoning stacks up against other frontier models in terms of cost-to-performance and raw intelligence (LMArena Elo).

2026 Performance Summary

🚀 Intelligence (Elo): 1472
💰 API Cost (Input): $0.00000125 / 1M tokens
⚡ API Cost (Output): $0.0000025 / 1M tokens
⏱️ Real-world Throughput: 92 tokens/s

Analysis: grok-4.20-beta-0309-reasoning offers competitive performance with an Elo of 1472. For developers optimizing strictly for cost-to-performance, we recommend comparing it against the current frontier leaders.

Highly Recommended Alternatives

Frontier Model	LMArena Elo	API Cost (1M)	Throughput
mimo-v2.5-pro	1467	$4.35e-7	50
gemini-3.5-flash-lite	1459	$3e-7	70
hy3	1453	$1.32e-7	74
gemma-4-31b	1451	$1e-7	145.5
qwen3-235b-a22b-instruct-2507	1423	$9e-8	50

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