Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | MiniMax M1 80K | o3-mini |
|---|---|---|
| GPQA Diamond | 70 | 77.2 |
| Humanity's Last Exam | 8.4 | — |
math
| Benchmark | MiniMax M1 80K | o3-mini |
|---|---|---|
| AIME 2025 | 76.9 | — |
| FrontierMath | — | 9.2 |
coding
| Benchmark | MiniMax M1 80K | o3-mini |
|---|---|---|
| SWE-bench Verified | 56 | 49.3 |
general
| Benchmark | MiniMax M1 80K | o3-mini |
|---|---|---|
| SimpleQA | 18.5 | 15 |
agent
| Benchmark | MiniMax M1 80K | o3-mini |
|---|---|---|
| TAU-Bench Retail | 63.5 | 57.6 |
Category Winners
coding
28.3
math
45.7
reasoning
68.4
general
22.4
agent
37.2