Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | GPT-4.1 | MiniMax M1 40K |
|---|---|---|
| GPQA Diamond | 66.3 | 69.2 |
| Humanity's Last Exam | 5.4 | 7.2 |
math
| Benchmark | GPT-4.1 | MiniMax M1 40K |
|---|---|---|
| AIME 2025 | 46.4 | 74.6 |
coding
| Benchmark | GPT-4.1 | MiniMax M1 40K |
|---|---|---|
| SWE-bench Verified | 54.6 | 55.6 |
multimodal
| Benchmark | GPT-4.1 | MiniMax M1 40K |
|---|---|---|
| MMMU | 74.8 | — |
| CharXiv Reasoning | 56.7 | — |
agent
| Benchmark | GPT-4.1 | MiniMax M1 40K |
|---|---|---|
| TAU-Bench Retail | 68 | 67.8 |
Category Winners
coding
26.4
math
41.1
reasoning
30.3
general
66.3
multimodal
38.9
agent
50.7