Stick To Your Role! - Rank-Order Stability: leaderboard
Metric: Rank-Order Stability (correlation, -1 to 1). Source: huggingface.co. Saturation forecast: Estimated already saturated. 31 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 2.5 VL 72B Instruct | 0.81 |
| 2 | Llama 3.3 70B Instruct | 0.81 |
| 3 | QwQ-32B | 0.81 |
| 4 | Llama 3.1 Nemotron 70B Instruct | 0.8 |
| 5 | Mistral Large 2 (Jul) | 0.78 |
| 6 | Llama 3.1 70B Instruct | 0.76 |
| 7 | Qwen 3 32B | 0.74 |
| 8 | Qwen 3 8B | 0.72 |
| 9 | Mistral Large 2 (Nov) Instruct (2411) | 0.71 |
| 10 | Mistral Small 3.1 | 0.69 |
| 11 | Qwen 3 235B A22B FP8 | 0.68 |
| 12 | Qwen 3 4B | 0.66 |
| 13 | Qwen2.5-14B-Instruct-1M | 0.66 |
| 14 | Llama 3.1 8B Instruct | 0.55 |
| 15 | GLM-4 32B (0414) | 0.52 |
Interactive version: theaggregate.ai/benchmark?slug=stick-to-your-role-rank-order-stability · How It Works · Data refreshed daily, snapshot 2026-09-26.