Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | Gemini 3.1 Flash-Lite | LongCat-Flash-Thinking |
|---|---|---|
| GPQA Diamond | 86.9 | 81.5 |
| Humanity's Last Exam | 16 | — |
multimodal
| Benchmark | Gemini 3.1 Flash-Lite | LongCat-Flash-Thinking |
|---|---|---|
| MMMU-Pro | 76.8 | — |
| CharXiv Reasoning | 73.2 | — |
math
| Benchmark | Gemini 3.1 Flash-Lite | LongCat-Flash-Thinking |
|---|---|---|
| AIME 2025 | — | 90.6 |
coding
| Benchmark | Gemini 3.1 Flash-Lite | LongCat-Flash-Thinking |
|---|---|---|
| SWE-bench Verified | — | 59.4 |
Category Winners
coding
M
LongCat-Flash-Thinking36.3
math
M
LongCat-Flash-Thinking70.8
reasoning
M
LongCat-Flash-Thinking77.1
general
68.2
multimodal
63.0