Harvey LAB-AA - Environmental & ESG: leaderboard
Metric: Criterion Pass Rate (%). Source: artificialanalysis.ai. 43 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Muse Spark 1.3 (xHigh) | 95.7 |
| 2 | Kimi K3 (Max) | 92.11 |
| 3 | Claude Opus 5 (Adaptive Reasoning, Max Effort) | 90.32 |
| 4 | Gemini 3.7 Flash (High) | 88.17 |
| 5 | Claude Opus 4.8 (Adaptive Reasoning, Max Effort) | 88.17 |
| 6 | Claude Sonnet 5 (Adaptive Reasoning, Max Effort) | 87.81 |
| 7 | MiniMax-M3 | 87.1 |
| 8 | GLM-5.2 (Max) | 87.1 |
| 9 | Grok 4.5 (High) | 86.02 |
| 10 | GPT-5.5 (xHigh) | 84.59 |
| 11 | Gemini 3.6 Flash (High) | 84.59 |
| 12 | Kimi K2.7 Code | 83.87 |
| 13 | GPT-5.6 Luna (Max) | 83.15 |
| 14 | GPT-5.6 Sol (Max) | 82.08 |
| 15 | Qwen 3.7 Max | 81.36 |
Interactive version: theaggregate.ai/benchmark?slug=harvey-lab-aa-environmental-esg · How It Works · Data refreshed daily, snapshot 2026-09-19.