BrickBench (Overall, BrickAgent) - Alignment Elo: leaderboard
Metric: Alignment Elo (rating); published Overall cohort. Source: www.brickben.ch. Saturation forecast: Around February 2028. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-6.1 Sol | 1301 |
| 2 | GPT-6 Astra | 1295 |
| 3 | Claude Opus 5.5 | 1212 |
| 4 | Claude Opus 5 | 1095 |
| 5 | GPT-5.6 Sol | 1041 |
| 6 | Qwen 3.8 Flash | 1040 |
| 7 | Gemini 3.8 Flash | 1005 |
| 8 | DeepSeek V4.1 Flash | 998 |
| 9 | GPT-5.6 Luna | 917 |
| 10 | Muse Spark 1.3 | 859 |
| 11 | GLM-5.3 Flash | 750 |
Interactive version: theaggregate.ai/benchmark?slug=brickbench-overall-brickagent-alignment-elo · How It Works · Data refreshed daily, snapshot 2026-10-10.