Generative Engine Optimization on GEO-bench 1,000 queries (test)
11.07Word ScoreIF-GEO
Evaluation Results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| IF-GEOUnderlying LLM=GPT-4o-mini, query expansion size Nq=5, suggestions per query Ns=5, internal temperature=0.2, global priority filtering threshold tau=0.72026.01 | 11.07 | 11.15 | 11.03 | 5.12 | 6.16 | 5.9 | 5.29 | 5.34 | 5.98 | 5.87 | |
| Auto-GEOUnderlying LLM=GPT-4o-mini2026.01 | 7.8 | 7.64 | 7.59 | 4.99 | 5.68 | 5.18 | 4.74 | 4.19 | 4.77 | 5.3 | |
| Cite SourcesUnderlying LLM=GPT-4o-mini2026.01 | 4.47 | 4.59 | 4.71 | 3.17 | 3.45 | 3.36 | 3.16 | 2.96 | 3.26 | 3.31 | |
| Quot. Addi.Underlying LLM=GPT-4o-mini2026.01 | 4.29 | 4.19 | 4.23 | 2.57 | 2.87 | 3.1 | 2.45 | 2.25 | 2.48 | 2.71 | |
| Stat. Addi.Underlying LLM=GPT-4o-mini2026.01 | 3.28 | 3.39 | 3.49 | 2.15 | 2.48 | 2.54 | 1.97 | 1.89 | 2.16 | 2.31 | |
| Tran. SEOUnderlying LLM=GPT-4o-mini2026.01 | 1.83 | 1.77 | 1.84 | 1.44 | 1.62 | 1.24 | 1.5 | 1.44 | 1.58 | 1.51 | |
| RAIDUnderlying LLM=GPT-4o-mini2026.01 | 1.06 | 0.78 | 0.88 | 1.44 | 1.75 | 1.4 | 1.08 | 0.84 | 1.34 | 1.36 | |
| Term. Addi.Underlying LLM=GPT-4o-mini2026.01 | 1 | 1.07 | 1.17 | 0.71 | 1.22 | 0.5 | 0.73 | 0.43 | 0.63 | 0.75 | |
| Auth. Expr.Underlying LLM=GPT-4o-mini2026.01 | 0.97 | 0.88 | 0.92 | 0.56 | 1.08 | 0.6 | 0.86 | 0.57 | 0.74 | 0.81 | |
| Flue. Expr.Underlying LLM=GPT-4o-mini2026.01 | 0.92 | 0.94 | 1.03 | 0.52 | 0.82 | 0.48 | 0.48 | 0.29 | 0.56 | 0.63 | |
| Simp. Expr.Underlying LLM=GPT-4o-mini2026.01 | 0.16 | 0.28 | 0.28 | 0.59 | 0.84 | 0.73 | 0.72 | 0.35 | 0.57 | 0.69 | |
| Uniq. WordUnderlying LLM=GPT-4o-mini2026.01 | -0.7 | -1 | -0.79 | -0.2 | -0.16 | -0.44 | -0.42 | -0.46 | -0.38 | -0.39 |