Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | GPT OSS 20B | Magistral Medium |
|---|---|---|
| GPQA Diamond | 71.5 | 70.8 |
| Humanity's Last Exam | 10.9 | 9 |
agent
| Benchmark | GPT OSS 20B | Magistral Medium |
|---|---|---|
| TAU-Bench Retail | 54.8 | — |
math
| Benchmark | GPT OSS 20B | Magistral Medium |
|---|---|---|
| AIME 2025 | — | 64.9 |
Category Winners
math
24.6
reasoning
34.9
agent
16.1