Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | DeepSeek-V3 0324 | Phi 4 Reasoning Plus |
|---|---|---|
| GPQA Diamond | 68.4 | 68.9 |
math
| Benchmark | DeepSeek-V3 0324 | Phi 4 Reasoning Plus |
|---|---|---|
| AIME 2025 | — | 78 |
Category Winners
math
47.8
reasoning
52.5