Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
language
| Benchmark | Claude 3 Opus | Muse Glimmer-30B |
|---|---|---|
| WinoGrande | 85.4 | — |
| HellaSwag | 95.4 | — |
general
| Benchmark | Claude 3 Opus | Muse Glimmer-30B |
|---|---|---|
| Chatbot Arena Elo | 1232 | — |
| IFEval | 85.2 | — |
| MMLU-Pro | 68.5 | — |
reasoning
| Benchmark | Claude 3 Opus | Muse Glimmer-30B |
|---|---|---|
| GPQA Diamond | 50.4 | 83.5 |
| BigBench-Hard | 86.8 | — |
| ARC-Challenge | 96.4 | — |
| Humanity's Last Exam | — | 22 |
safety
| Benchmark | Claude 3 Opus | Muse Glimmer-30B |
|---|---|---|
| TruthfulQA | 72.8 | — |
coding
| Benchmark | Claude 3 Opus | Muse Glimmer-30B |
|---|---|---|
| HumanEval | 84.9 | — |
| SWE-bench Verified | — | 76 |
| SWE-Bench Pro | — | 51.2 |
| SciCode | — | 43.6 |
multimodal
| Benchmark | Claude 3 Opus | Muse Glimmer-30B |
|---|---|---|
| MMMU-Pro | — | 74 |
| CharXiv Reasoning | — | 78.8 |
| ScreenSpot Pro | — | 75.4 |
agent
| Benchmark | Claude 3 Opus | Muse Glimmer-30B |
|---|---|---|
| MCP Atlas | — | 75.5 |
Category Winners
coding
51.2
math
44.3
reasoning
M
Muse Glimmer-30B54.8
general
56.6
language
75.0
multimodal
M
Muse Glimmer-30B69.8
safety
92.9
agent
M
Muse Glimmer-30B