Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
language
| Benchmark | Claude 3 Opus | Mistral Large 2 |
|---|---|---|
| WinoGrande | 85.4 | — |
| HellaSwag | 95.4 | 93.6 |
general
| Benchmark | Claude 3 Opus | Mistral Large 2 |
|---|---|---|
| MMLU-Pro | 68.5 | 69.4 |
| IFEval | 85.2 | 82.8 |
| Chatbot Arena Elo | 1232 | 1250 |
reasoning
| Benchmark | Claude 3 Opus | Mistral Large 2 |
|---|---|---|
| GPQA Diamond | 50.4 | 48.2 |
| BigBench-Hard | 86.8 | — |
| ARC-Challenge | 96.4 | 94 |
safety
| Benchmark | Claude 3 Opus | Mistral Large 2 |
|---|---|---|
| TruthfulQA | 72.8 | 60.5 |
coding
| Benchmark | Claude 3 Opus | Mistral Large 2 |
|---|---|---|
| HumanEval | 84.9 | 92 |
Category Winners
coding
80.3
math
46.2
reasoning
51.6
general
56.6
language
75.0
safety
92.9