Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | GPT-5.6 Luna | o1 |
|---|---|---|
| GPQA Diamond | 92.3 | 78 |
| BigBench-Hard | — | 91.5 |
coding
| Benchmark | GPT-5.6 Luna | o1 |
|---|---|---|
| SWE-Bench Pro | 62.7 | — |
| HumanEval | — | 92.4 |
| SWE-bench Verified | — | 48.9 |
multimodal
| Benchmark | GPT-5.6 Luna | o1 |
|---|---|---|
| MMMU-Pro | 78.4 | — |
agent
| Benchmark | GPT-5.6 Luna | o1 |
|---|---|---|
| BrowseComp | 83.3 | — |
| Toolathlon | 53.4 | — |
math
| Benchmark | GPT-5.6 Luna | o1 |
|---|---|---|
| FrontierMath | 78.6 | — |
| GSM8K | — | 97.8 |
| AIME 2024 | — | 83.3 |
| MATH | — | 96.4 |
general
| Benchmark | GPT-5.6 Luna | o1 |
|---|---|---|
| MRCR v2 | 41.3 | — |
| IFEval | — | 86.3 |
| Chatbot Arena Elo | — | 1350 |
| MMLU-Pro | — | 79.2 |
Category Winners
coding
73.6
math
95.5
reasoning
89.2
general
86.7
multimodal
77.6
agent
65.8