Mental Health Reasoning and Assessment on MentraBench 1.0 (test)
75.16Cognitive ReframingDeepseek-R1
Evaluation Results
| Method | Links | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Deepseek-R1Model Category=close-source models2025.12 | 75.16 | 70.69 | 44.72 | 63.52 | 40.85 | 76.24 | 72.18 | 63.09 | 20.2 | 49.84 | 86.08 | 89.88 | 72.56 | 74.59 | 90.37 | 56.89 | 65.05 | |
| Mindora CHORDModel Category=7-8B open-source models2025.12 | 72.93 | 68.42 | 50.88 | 64.08 | 46.55 | 77.6 | 80.3 | 68.15 | 40.16 | 63.17 | 75.9 | 85.35 | 84.42 | 77.21 | 83.79 | 71.78 | 69.33 | |
| GPT-o4-miniModel Category=close-source models2025.12 | 70.02 | 66.07 | 44.68 | 60.26 | 36.07 | 73.74 | 75.15 | 61.65 | 20.86 | 54.62 | 89.25 | 81.02 | 71.18 | 74.02 | 86.5 | 77.81 | 65.15 | |
| Qwen2.5-72BModel Category=70B+ open-source models2025.12 | 68.52 | 62.07 | 39.3 | 56.63 | 38.06 | 76.39 | 81.36 | 65.27 | 27.96 | 40.22 | 74.49 | 84.02 | 72.29 | 67.75 | 95.55 | 60.91 | 63.16 | |
| Qwen-plusModel Category=close-source models2025.12 | 68.21 | 63.95 | 40.57 | 57.58 | 40.43 | 76.24 | 76.62 | 64.43 | 18.66 | 47.95 | 82.42 | 88.95 | 71.99 | 72.83 | 90.48 | 60.18 | 63.87 | |
| GPT-4oModel Category=close-source models2025.12 | 67.91 | 66.67 | 45.43 | 60 | 40.64 | 75 | 71.14 | 62.26 | 33.58 | 45.31 | 61.38 | 79.36 | 69.49 | 63.89 | 90.65 | 68.94 | 62.73 | |
| QwQ-32BModel Category=32B open-source models2025.12 | 67.91 | 68.42 | 39.05 | 58.46 | 40.85 | 75 | 72.96 | 62.94 | 17.49 | 43.05 | 68.31 | 84.7 | 73.75 | 67.45 | 92.96 | 46.42 | 60.84 | |
| QwQ-plusModel Category=close-source models2025.12 | 66.98 | 68.71 | 36.98 | 57.56 | 40.85 | 75.47 | 73.3 | 63.21 | 18.48 | 42.01 | 64.05 | 83.54 | 72.26 | 65.46 | 93.35 | 48.38 | 60.34 | |
| dsdistill-LLaMA-70BModel Category=70B+ open-source models2025.12 | 66.67 | 66.67 | 39.05 | 57.46 | 40 | 75.16 | 69.53 | 61.56 | 23.67 | 44.04 | 70.97 | 79.57 | 74.52 | 67.27 | 89.14 | 68.14 | 62.09 | |
| dsdistill-Qwen14BModel Category=14B open-source models2025.12 | 66.35 | 59.81 | 34.04 | 53.4 | 41.69 | 74.21 | 77.5 | 64.47 | 24.7 | 38.23 | 61.98 | 66.6 | 74.97 | 60.45 | 86.41 | 57.95 | 58.8 | |
| LLaMA-4Model Category=70B+ open-source models2025.12 | 65.88 | 64.86 | 40.31 | 57.02 | 39.14 | 73.1 | 78.84 | 63.69 | 20.66 | 46.84 | 74.14 | 89 | 72.39 | 70.59 | 76.34 | 61.58 | 61.78 | |
| Deepseek-V3Model Category=close-source models2025.12 | 65.4 | 67.55 | 41.31 | 58.09 | 42.11 | 77.15 | 75.81 | 65.02 | 21.4 | 47.68 | 81.84 | 86.96 | 68.48 | 71.24 | 92.96 | 61.56 | 63.86 | |
| dsdistill-Qwen32BModel Category=32B open-source models2025.12 | 65.4 | 66.96 | 35.39 | 55.92 | 39.79 | 73.58 | 73.73 | 62.37 | 21.78 | 40.1 | 66.97 | 77.95 | 72.29 | 64.33 | 92.25 | 61.77 | 60.61 | |
| Qwen3-14BModel Category=14B open-source models2025.12 | 65.4 | 64.86 | 48.2 | 59.49 | 38.71 | 74.84 | 75.4 | 62.98 | 16.8 | 42.15 | 70.47 | 81.02 | 68.82 | 65.62 | 81.55 | 66 | 61.09 | |
| LLaMA-3.3-70BModel Category=70B+ open-source models2025.12 | 64.44 | 63.01 | 35.66 | 54.37 | 41.48 | 71.63 | 71.57 | 61.56 | 20.23 | 41.48 | 77.12 | 89.04 | 72.99 | 70.16 | 63.18 | 68.94 | 60.06 | |
| Qwen3-32BModel Category=32B open-source models2025.12 | 62.47 | 63.95 | 42.55 | 56.32 | 38.06 | 76.85 | 80.38 | 65.1 | 20.16 | 43.55 | 74.15 | 82.06 | 70.83 | 67.65 | 93.33 | 50.25 | 61.43 | |
| EmoLLMModel Category=7-8B open-source models2025.12 | 61.8 | 54.72 | 33.22 | 49.91 | 43.34 | 72.29 | 79.31 | 64.98 | 14.08 | 33.73 | 43.11 | 57.49 | 67.35 | 50.42 | 79.06 | 63.46 | 54.07 | |
| Mindora SFT+RLModel Category=7-8B open-source models2025.12 | 59.75 | 62.07 | 45.66 | 55.83 | 52.4 | 77.15 | 81.4 | 70.32 | 38.03 | 59.96 | 78.39 | 85.93 | 82.12 | 76.6 | 71.59 | 56.81 | 65.48 | |
| Qwen3-8BModel Category=7-8B open-source models2025.12 | 59.41 | 63.95 | 40.57 | 54.64 | 41.69 | 73.58 | 75.15 | 63.47 | 14.89 | 40.09 | 58.24 | 79.27 | 68.99 | 61.65 | 81.15 | 47.73 | 57.29 | |
| dsdistill-LLaMA-8BModel Category=7-8B open-source models2025.12 | 59.06 | 55.07 | 23.96 | 46.03 | 41.9 | 72.94 | 74.9 | 63.25 | 20.01 | 33.21 | 44.3 | 60.7 | 75.87 | 53.52 | 81.57 | 76.17 | 55.36 | |
| LLaMA3.1-8BModel Category=7-8B open-source models2025.12 | 58.71 | 57.14 | 19.57 | 45.14 | 44.76 | 73.58 | 75.81 | 64.72 | 21.47 | 27.71 | 57.91 | 79.25 | 70.96 | 58.96 | 68.05 | 67.03 | 55.53 | |
| Mindora SFTModel Category=7-8B open-source models2025.12 | 56.93 | 62.07 | 44.72 | 54.57 | 48.87 | 74.37 | 75.97 | 66.4 | 41.93 | 58.3 | 72.63 | 85.35 | 73.76 | 72.51 | 77.64 | 55.12 | 63.67 | |
| Psyche-R1Model Category=7-8B open-source models2025.12 | 50.13 | 52.22 | 20.21 | 40.85 | 41.69 | 68.77 | 79.62 | 63.36 | 24.24 | 58.72 | 69.89 | 84.01 | 71.94 | 71.14 | 81.64 | 69.54 | 59.43 |