Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | Gemini 2.5 Pro | Mistral Large 2 |
|---|---|---|
| GPQA Diamond | 83 | 48.2 |
| Humanity's Last Exam | 17.8 | — |
| ARC-AGI v2 | 4.9 | — |
| ARC-Challenge | — | 94 |
coding
| Benchmark | Gemini 2.5 Pro | Mistral Large 2 |
|---|---|---|
| SWE-bench Verified | 63.2 | — |
| HumanEval | — | 92 |
multimodal
| Benchmark | Gemini 2.5 Pro | Mistral Large 2 |
|---|---|---|
| MMMU | 79.6 | — |
general
| Benchmark | Gemini 2.5 Pro | Mistral Large 2 |
|---|---|---|
| SimpleQA | 50.8 | — |
| Chatbot Arena Elo | — | 1250 |
| MMLU-Pro | — | 69.4 |
| IFEval | — | 82.8 |
language
| Benchmark | Gemini 2.5 Pro | Mistral Large 2 |
|---|---|---|
| HellaSwag | — | 93.6 |
safety
| Benchmark | Gemini 2.5 Pro | Mistral Large 2 |
|---|---|---|
| TruthfulQA | — | 60.5 |
Category Winners
coding
80.3
math
57.4
reasoning
36.5
general
68.7
language
66.8
multimodal
80.3
safety
29.8