Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | Claude 3.5 Haiku | GPT-4o mini |
|---|---|---|
| GPQA Diamond | 41.6 | 40.2 |
coding
| Benchmark | Claude 3.5 Haiku | GPT-4o mini |
|---|---|---|
| SWE-bench Verified | 40.6 | 8.7 |
agent
| Benchmark | Claude 3.5 Haiku | GPT-4o mini |
|---|---|---|
| TAU-Bench Retail | 51 | — |
multimodal
| Benchmark | Claude 3.5 Haiku | GPT-4o mini |
|---|---|---|
| MMMU | — | 59.4 |
Category Winners
coding
7.0
reasoning
10.4
multimodal
28.4
agent
10.1