BrainBench (EEG) - BrainAgent - Neurocognitive Assessment: leaderboard
BrainBench (arXiv 2608.04156v2, Table 1) score on the Neurocognitive Assessment subset built on cognitive-experiment EEG, with each model running through the structured BrainAgent tool workflow in an isolated container and its report and artifacts checked by task validators. 0-100; higher is better. Optional reasoning modes disabled.
Metric: EEG analysis score (0-100). Source: arxiv.org. Saturation forecast: Around March 2027. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.6 Sol (Non-reasoning) | 72.61 | #22 (GPT-5.6 Sol) |
| 2 | GPT-5.6 Terra (Non-reasoning) | 72.16 | #40 (GPT-5.6 Terra) |
| 3 | Claude Opus 5 | 71.32 | #16 |
| 4 | Qwen 3.7 Max (Non-reasoning) | 69.94 | #85 (Qwen 3.7 Max) |
| 5 | Qwen 3.7 Plus (Non-reasoning) | 69.86 | #98 (Qwen 3.7 Plus) |
| 6 | DeepSeek V4 Flash (Non-reasoning) | 67.21 | #155 (DeepSeek V4 Flash) |
| 7 | Kimi K2.5 (Non-reasoning) | 65.56 | #152 (Kimi K2.5) |
| 8 | GPT-5.6 Luna (Non-reasoning) | 64.89 | #79 (GPT-5.6 Luna) |
| 9 | MiniMax-M2.5 | 63.81 | #316 |
| 10 | GLM-5 (Non-reasoning) | 63.56 | #149 (GLM-5) |
| 11 | Qwen 3.5 122B A10B (Non-reasoning) | 58.51 | #184 (Qwen 3.5 122B A10B) |
| 12 | Qwen 3.5 35B A3B (Non-reasoning) | 51.5 | #268 (Qwen 3.5 35B A3B) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=brainbench-eeg-brainagent-neurocognitive-assessment · How It Works · Data refreshed daily, snapshot 2026-11-05.