Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | GPT-5 | LongCat-Flash-Thinking-2601 |
|---|---|---|
| GPQA Diamond | 85.7 | 80.5 |
| Humanity's Last Exam | 24.8 | 25.2 |
math
| Benchmark | GPT-5 | LongCat-Flash-Thinking-2601 |
|---|---|---|
| AIME 2025 | 94.6 | 99.6 |
| FrontierMath | 26.3 | — |
coding
| Benchmark | GPT-5 | LongCat-Flash-Thinking-2601 |
|---|---|---|
| SWE-bench Verified | 74.9 | 70 |
multimodal
| Benchmark | GPT-5 | LongCat-Flash-Thinking-2601 |
|---|---|---|
| MMMU | 84.2 | — |
| MMMU-Pro | 78.4 | — |
| CharXiv Reasoning | 81.1 | — |
agent
| Benchmark | GPT-5 | LongCat-Flash-Thinking-2601 |
|---|---|---|
| BrowseComp | 54.9 | 56.6 |
Category Winners
coding
71.8
math
M
LongCat-Flash-Thinking-260183.5
reasoning
60.3
multimodal
79.2
agent
M
LongCat-Flash-Thinking-260134.6