Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | GPT-4o mini | LFM2.5-2.6B |
|---|---|---|
| GPQA Diamond | 40.2 | — |
coding
| Benchmark | GPT-4o mini | LFM2.5-2.6B |
|---|---|---|
| SWE-bench Verified | 8.7 | — |
multimodal
| Benchmark | GPT-4o mini | LFM2.5-2.6B |
|---|---|---|
| MMMU | 59.4 | — |
math
| Benchmark | GPT-4o mini | LFM2.5-2.6B |
|---|---|---|
| AIME 2025 | — | 51.9 |
Category Winners
coding
0.0
math
L
LFM2.5-2.6B9.5
reasoning
8.4
multimodal
27.9