Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | DeepSeek-R1-0528 | LongCat-Flash-Chat |
|---|---|---|
| GPQA Diamond | 81 | 73.2 |
| Humanity's Last Exam | 17.7 | — |
math
| Benchmark | DeepSeek-R1-0528 | LongCat-Flash-Chat |
|---|---|---|
| AIME 2025 | 87.5 | 61.3 |
coding
| Benchmark | DeepSeek-R1-0528 | LongCat-Flash-Chat |
|---|---|---|
| SWE-bench Verified | 44.6 | 60.4 |
| TerminalBench | 5.7 | 39.5 |
general
| Benchmark | DeepSeek-R1-0528 | LongCat-Flash-Chat |
|---|---|---|
| SimpleQA | 92.3 | — |
agent
| Benchmark | DeepSeek-R1-0528 | LongCat-Flash-Chat |
|---|---|---|
| BrowseComp | 8.9 | — |
Category Winners
coding
M
LongCat-Flash-Chat52.8
math
65.6
reasoning
M
LongCat-Flash-Chat61.0
general
98.3
agent
0.3