Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
coding
| Benchmark | Devin | Phi-3.5-MoE-instruct |
|---|---|---|
| SWE-bench Verified | 26.6 | — |
reasoning
| Benchmark | Devin | Phi-3.5-MoE-instruct |
|---|---|---|
| GPQA Diamond | — | 36.8 |
general
| Benchmark | Devin | Phi-3.5-MoE-instruct |
|---|---|---|
| MMMLU | — | 69.9 |
Category Winners
coding
0.9
reasoning
6.1
general
8.4
agent
10.0