Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | Claude 3.5 Sonnet | Magistral Small 2506 |
|---|---|---|
| GPQA Diamond | 67.2 | 68.2 |
coding
| Benchmark | Claude 3.5 Sonnet | Magistral Small 2506 |
|---|---|---|
| SWE-bench Verified | 49 | — |
multimodal
| Benchmark | Claude 3.5 Sonnet | Magistral Small 2506 |
|---|---|---|
| MMMU | 68.3 | — |
agent
| Benchmark | Claude 3.5 Sonnet | Magistral Small 2506 |
|---|---|---|
| TAU-Bench Retail | 69.2 | — |
math
| Benchmark | Claude 3.5 Sonnet | Magistral Small 2506 |
|---|---|---|
| AIME 2025 | — | 62.8 |
Category Winners
coding
15.0
math
21.1
reasoning
50.2
multimodal
50.6
agent
54.3