Performance Benchmarks for Gemma 3 27B IT
Analyze how Gemma 3 27B IT stacks up against other frontier models in terms of cost-to-performance and raw intelligence (LMArena Elo).
2026 Performance Summary
- 🚀 Intelligence (Elo): 1365
- 💰 API Cost (Input): $8e-8 / 1M tokens
- ⚡ API Cost (Output): $4.5e-7 / 1M tokens
- ⏱️ Real-world Throughput: 38 tokens/s
Analysis: While Gemma 3 27B IT is a capable model, it is currently outperformed in efficiency by Qwen4-Max (xHigh). Qwen4-Max (xHigh) sits higher on the Pareto line, offering approximately 385 points more intelligence (Elo) for a comparable or lower API cost.
Highly Recommended Alternatives
*These models represent the Pareto Frontier (optimal cost-to-performance).*