Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | DeepSeek R1 Distill Qwen 14B | Gemini 3.5 Flash-Lite |
|---|---|---|
| GPQA Diamond | 59.1 | — |
coding
| Benchmark | DeepSeek R1 Distill Qwen 14B | Gemini 3.5 Flash-Lite |
|---|---|---|
| SWE-Bench Pro | — | 54.2 |
multimodal
| Benchmark | DeepSeek R1 Distill Qwen 14B | Gemini 3.5 Flash-Lite |
|---|---|---|
| CharXiv Reasoning | — | 76.5 |
general
| Benchmark | DeepSeek R1 Distill Qwen 14B | Gemini 3.5 Flash-Lite |
|---|---|---|
| MRCR v2 | — | 21.3 |
Category Winners
coding
37.6
reasoning
33.4
general
15.9
multimodal
54.2