Neura Intelligence Index
| Benchmark | Grok-4 | LongCat-Flash-Thinking-2601 |
|---|---|---|
| GPQA Diamond | 87.5 | 80.5 |
| Humanity's Last Exam | 40 | 25.2 |
| ARC-AGI v2 | 15.9 | β |
| Benchmark | Grok-4 | LongCat-Flash-Thinking-2601 |
|---|---|---|
| AIME 2025 | 91.7 | 99.6 |
| Benchmark | Grok-4 | LongCat-Flash-Thinking-2601 |
|---|---|---|
| SWE-bench Verified | β | 70 |
| Benchmark | Grok-4 | LongCat-Flash-Thinking-2601 |
|---|---|---|
| BrowseComp | β | 56.6 |