Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | Llama 3.1 Nemotron Ultra 253B v1 | Nova 2 Lite |
|---|---|---|
| GPQA Diamond | 76.0 | 79.6 |
math
| Benchmark | Llama 3.1 Nemotron Ultra 253B v1 | Nova 2 Lite |
|---|---|---|
| AIME 2025 | 72.5 | 91 |
coding
| Benchmark | Llama 3.1 Nemotron Ultra 253B v1 | Nova 2 Lite |
|---|---|---|
| SWE-bench Verified | — | 64.5 |
| TerminalBench | — | 32.5 |
multimodal
| Benchmark | Llama 3.1 Nemotron Ultra 253B v1 | Nova 2 Lite |
|---|---|---|
| MMMU-Pro | — | 61.8 |
agent
| Benchmark | Llama 3.1 Nemotron Ultra 253B v1 | Nova 2 Lite |
|---|---|---|
| MCP Atlas | — | 24.6 |
Category Winners
coding
Nova 2 Lite44.9
math
Nova 2 Lite71.5
reasoning
Nova 2 Lite74.1
multimodal
Nova 2 Lite32.3
agent
Nova 2 Lite0.0