Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | o4-mini | Qwen3-235B-A22B-Thinking-2507 |
|---|---|---|
| GPQA Diamond | 81.4 | 81.1 |
| Humanity's Last Exam | 14.7 | 18.2 |
math
| Benchmark | o4-mini | Qwen3-235B-A22B-Thinking-2507 |
|---|---|---|
| AIME 2025 | 92.7 | 92.3 |
coding
| Benchmark | o4-mini | Qwen3-235B-A22B-Thinking-2507 |
|---|---|---|
| SWE-bench Verified | 68.1 | — |
multimodal
| Benchmark | o4-mini | Qwen3-235B-A22B-Thinking-2507 |
|---|---|---|
| MMMU | 81.6 | — |
| CharXiv Reasoning | 72 | — |
agent
| Benchmark | o4-mini | Qwen3-235B-A22B-Thinking-2507 |
|---|---|---|
| BrowseComp | 51.5 | — |
| TAU-Bench Retail | 71.8 | 67.8 |
Category Winners
coding
56.7
math
74.1
reasoning
Q
Qwen3-235B-A22B-Thinking-250750.5
multimodal
66.2
agent
Q
Qwen3-235B-A22B-Thinking-250750.1