Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
coding
| Benchmark | MiniMax M2.7 | Qwen3 235B A22B |
|---|---|---|
| SWE-Bench Pro | 56.2 | — |
agent
| Benchmark | MiniMax M2.7 | Qwen3 235B A22B |
|---|---|---|
| Toolathlon | 46.3 | — |
reasoning
| Benchmark | MiniMax M2.7 | Qwen3 235B A22B |
|---|---|---|
| GPQA Diamond | — | 47.5 |
math
| Benchmark | MiniMax M2.7 | Qwen3 235B A22B |
|---|---|---|
| AIME 2025 | — | 81.5 |
general
| Benchmark | MiniMax M2.7 | Qwen3 235B A22B |
|---|---|---|
| MMMLU | — | 86.7 |
Category Winners
coding
46.3
math
Q
Qwen3 235B A22B54.5
reasoning
Q
Qwen3 235B A22B15.6
general
Q
Qwen3 235B A22B63.7
agent
29.7