Grounding on ScreenSpot Pro
51.8Average Grounding AccuracyFull Cache
Evaluation Results
| Method | Links | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Full CacheModel=OpenCUA-7B, KV-cache budget=Full Cache2026.06 | 51.8 | 46.81 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| STaR-KVModel=OpenCUA-7B, KV-cache budget=40%2026.06 | 50.04 | 46.68 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| STaR-KVModel=UI-TARS-1.5-7B, KV-cache budget=40%2026.06 | 49.94 | 41.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Full CacheModel=UI-TARS-1.5-7B, KV-cache budget=Full Cache2026.06 | 49.75 | 41.68 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PyramidKVModel=OpenCUA-7B, KV-cache budget=40%2026.06 | 49.47 | 45.86 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SnapKVModel=OpenCUA-7B, KV-cache budget=40%2026.06 | 49.32 | 44.28 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUIKVModel=UI-TARS-1.5-7B, KV-cache budget=40%2026.06 | 48.92 | 40.73 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UI-AGILE-7B + Decomposed GroundingProtocol=Zero Shot / Reinforcement Fine-tuning, Examples=9K, Epochs=2, Decomposed Grounding=true2025.07 | 48.7 | — | 79.1 | 24.1 | 60.6 | 11.2 | 53.3 | 10.9 | 66 | 26.4 | 79.1 | 39.6 | 59.8 | 22.5 | 59.2 | |
| STaR-KVModel=UI-TARS-1.5-7B, KV-cache budget=20%2026.06 | 47.31 | 40.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUIKVModel=OpenCUA-7B, KV-cache budget=40%2026.06 | 47.31 | 44.28 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SnapKVModel=UI-TARS-1.5-7B, KV-cache budget=40%2026.06 | 47.22 | 41.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| STaR-KVModel=OpenCUA-7B, KV-cache budget=20%2026.06 | 47.01 | 45.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUIKVModel=UI-TARS-1.5-7B, KV-cache budget=20%2026.06 | 46.59 | 39.85 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PyramidKVModel=UI-TARS-1.5-7B, KV-cache budget=40%2026.06 | 45.45 | 38.71 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PyramidKVModel=OpenCUA-7B, KV-cache budget=20%2026.06 | 45.3 | 42.57 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SnapKVModel=UI-TARS-1.5-7B, KV-cache budget=20%2026.06 | 45.09 | 39.22 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UI-AGILE-3B + Decomposed GroundingProtocol=Zero Shot / Reinforcement Fine-tuning, Examples=9K, Epochs=2, Decomposed Grounding=true2025.07 | 45 | — | 66.9 | 16.6 | 58.1 | 11.9 | 47.2 | 10.9 | 66.7 | 24.5 | 72.3 | 34 | 58.9 | 22.5 | 54.4 | |
| UI-AGILE-7BProtocol=Zero Shot / Reinforcement Fine-tuning, Examples=9K, Epochs=22025.07 | 44 | — | 64.3 | 15.2 | 53 | 9.8 | 49.2 | 14.1 | 72.9 | 25.5 | 75.1 | 30.2 | 45.8 | 20.2 | — | |
| UI-R1-E + Decomposed GroundingProtocol=Zero Shot / Reinforcement Fine-tuning, Examples=2K, Epochs=8, Decomposed Grounding=true2025.07 | 43.3 | — | 63.6 | 17.2 | 59.6 | 10 | 43.7 | 6.3 | 66 | 21.8 | 68.4 | 43.4 | 56.1 | 19.1 | 52.6 | |
| GUIKVModel=OpenCUA-7B, KV-cache budget=20%2026.06 | 43.03 | 40.04 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| STaR-KVModel=OpenCUA-7B, KV-cache budget=10%2026.06 | 42.67 | 44.21 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UI-TARS-7B + Decomposed GroundingProtocol=Supervised Fine-tuning, Decomposed Grounding=true2025.07 | 41.9 | — | 59.7 | 19.3 | 54 | 15.4 | 38.1 | 12.5 | 63.2 | 27.3 | 71.8 | 28.3 | 45.8 | 21.3 | 50.3 | |
| SnapKVModel=OpenCUA-7B, KV-cache budget=20%2026.06 | 41.17 | 45.48 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| STaR-KVModel=UI-TARS-1.5-7B, KV-cache budget=10%2026.06 | 40.72 | 37.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUIKVModel=UI-TARS-1.5-7B, KV-cache budget=10%2026.06 | 40.7 | 36.05 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SnapKVModel=UI-TARS-1.5-7B, KV-cache budget=10%2026.06 | 40.04 | 35.74 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PyramidKVModel=OpenCUA-7B, KV-cache budget=10%2026.06 | 39.65 | 38.96 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| JEDI-7BProtocol=Supervised Fine-tuning, Examples=4M2025.07 | 39.5 | — | 42.9 | 11 | 50 | 11.9 | 38 | 14.1 | 72.9 | 25.5 | 75.1 | 47.2 | 33.6 | 16.9 | — | |
| PyramidKVModel=UI-TARS-1.5-7B, KV-cache budget=20%2026.06 | 39.46 | 35.36 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUI-R1-7B + Decomposed GroundingProtocol=Zero Shot / Reinforcement Fine-tuning, Examples=3K, Epochs=9, Decomposed Grounding=true2025.07 | 39.3 | — | 66.9 | 15.2 | 50 | 10.5 | 32.5 | 4.7 | 59 | 13.6 | 68.4 | 24.5 | 60.7 | 18 | 48.6 | |
| GUIKVModel=OpenCUA-7B, KV-cache budget=10%2026.06 | 39.16 | 37.82 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SnapKVModel=OpenCUA-7B, KV-cache budget=10%2026.06 | 38.27 | 44.09 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UI-TARS-72BProtocol=Supervised Fine-tuning2025.07 | 38.1 | — | 63 | 17.3 | 57.1 | 15.4 | 18.8 | 17.2 | 64.6 | 20.9 | 63.3 | 26.4 | 42.1 | 15.7 | — | |
| UI-AGILE-3BProtocol=Zero Shot / Reinforcement Fine-tuning, Examples=9K, Epochs=22025.07 | 37.9 | — | 53.2 | 9 | 50.5 | 8.4 | 44.2 | 20.3 | 62.5 | 22.7 | 65.5 | 22.6 | 35.5 | 12.4 | — | |
| RULERProtocol=Supervised Fine-tuning, Examples=8M, Epochs=12025.07 | 37.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUI-G1-3BProtocol=Zero Shot / Reinforcement Fine-tuning, Examples=17K, Epochs=12025.07 | 37.1 | — | 50.7 | 10.3 | 36.6 | 11.9 | 39.6 | 9.4 | 61.8 | 30 | 67.2 | 32.1 | 32.5 | 10.6 | — | |
| GUI-R1-3B + Decomposed GroundingProtocol=Zero Shot / Reinforcement Fine-tuning, Examples=3K, Epochs=9, Decomposed Grounding=true2025.07 | 37.1 | — | 63.6 | 13.1 | 55.6 | 4.9 | 31.5 | 6.3 | 61.8 | 16.4 | 62.7 | 20.8 | 44.9 | 10.1 | 46 | |
| UGround-V1-7B + Decomposed GroundingProtocol=Supervised Fine-tuning, Examples=10M, Decomposed Grounding=true2025.07 | 36.6 | — | 57.8 | 14.5 | 49 | 11.9 | 20.3 | 7.8 | 62.5 | 21.8 | 67.8 | 18.9 | 48.6 | 14.6 | 47.3 | |
| Aguvis-7B + Decomposed GroundingProtocol=Supervised Fine-tuning, Examples=1M, Epochs=1, Decomposed Grounding=true2025.07 | 36.5 | — | 50.6 | 11.7 | 60.1 | 7 | 31 | 4.7 | 62.5 | 20 | 63.3 | 18.9 | 45.8 | 6.7 | 44.3 | |
| JEDI-3BProtocol=Supervised Fine-tuning, Examples=4M2025.07 | 36.1 | — | 61 | 13.8 | 53.5 | 8.4 | 27.4 | 9.4 | 54.2 | 18.2 | 64.4 | 32.1 | 38.3 | 9 | — | |
| STaR-KVModel=OpenCUA-7B, KV-cache budget=5%2026.06 | 35.89 | 38.96 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UI-TARS-7BProtocol=Supervised Fine-tuning2025.07 | 35.7 | — | 58.4 | 12.4 | 50 | 9.1 | 20.8 | 9.4 | 63.9 | 31.8 | 63.3 | 20.8 | 30.8 | 16.9 | — | |
| InfiGUI-R1-3BProtocol=Zero Shot / Reinforcement Fine-tuning, Examples=32K2025.07 | 35.7 | — | 51.3 | 12.4 | 44.9 | 7 | 33 | 14.1 | 58.3 | 20 | 65.5 | 28.3 | 43.9 | 12.4 | — | |
| GUIKVModel=OpenCUA-7B, KV-cache budget=5%2026.06 | 33.78 | 32.95 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SnapKVModel=OpenCUA-7B, KV-cache budget=5%2026.06 | 33.62 | 38.77 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UI-R1-EProtocol=Zero Shot / Reinforcement Fine-tuning, Examples=2K, Epochs=82025.07 | 33.5 | — | 46.1 | 6.9 | 41.9 | 4.2 | 37.1 | 12.5 | 56.9 | 21.8 | 65 | 26.4 | 32.7 | 10.1 | — | |
| Qwen2.5-VL-7B + Decomposed GroundingProtocol=Zero Shot / Reinforcement Fine-tuning, Decomposed Grounding=true2025.07 | 33.3 | — | 60.4 | 13.1 | 33.3 | 8.4 | 27.9 | 6.2 | 50 | 13.6 | 63.3 | 17 | 51.4 | 16 | 42 | |
| PyramidKVModel=UI-TARS-1.5-7B, KV-cache budget=10%2026.06 | 33.12 | 29.48 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OS-Atlas-7B + Decomposed GroundingProtocol=Supervised Fine-tuning, Examples=13M, Decomposed Grounding=true2025.07 | 33.1 | — | 49.4 | 5.5 | 52 | 5.6 | 26.4 | 6.3 | 54.9 | 18.2 | 57.6 | 18.9 | 49.5 | 9 | 42.1 | |
| STaR-KVModel=UI-TARS-1.5-7B, KV-cache budget=5%2026.06 | 32.74 | 30.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PyramidKVModel=OpenCUA-7B, KV-cache budget=5%2026.06 | 32.72 | 33.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUIKVModel=UI-TARS-1.5-7B, KV-cache budget=5%2026.06 | 32.63 | 30.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUI-R1-7BProtocol=Zero Shot / Reinforcement Fine-tuning, Examples=3K, Epochs=92025.07 | 32.1 | — | 57.1 | 8.3 | 37.9 | 8.4 | 28.4 | 6.3 | 54.9 | 10.9 | 59.9 | 13.2 | 41.1 | 13.5 | — | |
| SnapKVModel=UI-TARS-1.5-7B, KV-cache budget=5%2026.06 | 32.04 | 29.22 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UGround-V1-7BProtocol=Supervised Fine-tuning, Examples=10M2025.07 | 31.6 | — | 51.3 | 5.5 | 48.5 | 8.3 | 18.8 | 1.6 | 59.7 | 14.6 | 59.9 | 17 | 40.2 | 7.9 | — | |
| Qwen2.5-VL-3B + Decomposed GroundingProtocol=Zero Shot / Reinforcement Fine-tuning, Decomposed Grounding=true2025.07 | 31.2 | — | 52.6 | 8.3 | 42.9 | 11.2 | 25.9 | 3.1 | 47.9 | 10 | 55.9 | 17 | 46.7 | 9 | 38.8 | |
| GUI-R1-3BProtocol=Zero Shot / Reinforcement Fine-tuning, Examples=3K, Epochs=92025.07 | 30.9 | — | 40.9 | 4.8 | 47.8 | 2.8 | 27.9 | 6.3 | 65.3 | 19.1 | 58.2 | 18.9 | 29 | 2.2 | — | |
| UI-TARS-2BProtocol=Supervised Fine-tuning2025.07 | 27.7 | — | 47.7 | 4.1 | 42.9 | 6.3 | 17.8 | 4.7 | 56.9 | 17.3 | 50.3 | 17 | 21.5 | 5.6 | — | |
| PyramidKVModel=UI-TARS-1.5-7B, KV-cache budget=5%2026.06 | 24.7 | 20.75 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen2.5-VL-7BProtocol=Zero Shot / Reinforcement Fine-tuning2025.07 | 24.5 | — | 54.5 | 5.5 | 24.7 | 4.2 | 13.7 | 3.1 | 46.5 | 7.3 | 50.8 | 11.3 | 29.9 | 10.1 | — | |
| Qwen2.5-VL-3BProtocol=Zero Shot / Reinforcement Fine-tuning2025.07 | 22.7 | — | 31.8 | 4.1 | 32.8 | 4.2 | 24.9 | 4.7 | 43.8 | 12.7 | 42.4 | 15.1 | 17.8 | 2.2 | — | |
| Aguvis-7BProtocol=Supervised Fine-tuning, Examples=1M, Epochs=12025.07 | 20.4 | — | 30.5 | 0.7 | 28.8 | 2.8 | 14.7 | 1.6 | 45.8 | 8.2 | 38.4 | 11.3 | 30.8 | 2.3 | — | |
| Qwen2.5-VL-7B + VPSGParadigm=Generalist Models or Training-free Methods, test-time guidance=VPSG2025.10 | 19.1 | — | 40.9 | 2.1 | 19.8 | 2.8 | 8.1 | 1.6 | 30.6 | 5.6 | 43.5 | 13.2 | 29.9 | 10.1 | — | |
| OS-Atlas-7BProtocol=Supervised Fine-tuning, Examples=13M2025.07 | 18.9 | — | 33.8 | 1.4 | 30.8 | 3.5 | 12.2 | 3.1 | 33.3 | 9.1 | 33.3 | 9.4 | 26.2 | 3.4 | — | |
| OS-Atlas-7BParadigm=Trained GUI Action Models2025.10 | 18.9 | — | 33.1 | 1.4 | 28.8 | 2.8 | 12.2 | 4.7 | 37.5 | 7.3 | 33.9 | 5.7 | 27.1 | 4.5 | — | |
| Qwen2.5-VL-7BParadigm=Generalist Models or Training-free Methods2025.10 | 18.5 | — | 37.7 | 2.8 | 19.7 | 2.1 | 7.6 | 1.6 | 31.3 | 5.5 | 41.8 | 11.3 | 29.9 | 10.1 | — | |
| UI-R1-3BProtocol=Zero Shot / Reinforcement Fine-tuning, Examples=136, Epochs=82025.07 | 17.8 | — | 22.7 | 4.1 | 27.3 | 3.5 | 11.2 | 6.3 | 42.4 | 11.8 | 32.2 | 11.3 | 13.1 | 4.5 | — | |
| UGround-7BParadigm=Trained GUI Action Models2025.10 | 16.5 | — | 26.6 | 2.1 | 27.3 | 2.8 | 14.2 | 1.6 | 31.9 | 2.7 | 31.6 | 11.3 | 17.8 | 0 | — | |
| Qwen2.5-VL-3B + VPSGParadigm=Generalist Models or Training-free Methods, test-time guidance=VPSG2025.10 | 13.3 | — | 24.7 | 2.1 | 20.2 | 2.1 | 10.2 | 1.6 | 21.5 | 5.5 | 26.6 | 5.7 | 15.9 | 1.1 | — | |
| Qwen2.5-VL-3BParadigm=Generalist Models or Training-free Methods2025.10 | 11.6 | — | 18.8 | 1.4 | 16.7 | 1.4 | 8.1 | 1.6 | 20.8 | 5.5 | 24.3 | 1.9 | 16.8 | 3.4 | — | |
| Aria-UIProtocol=Supervised Fine-tuning, Examples=16.6M2025.07 | 11.3 | — | 16.2 | 0 | 23.7 | 2.1 | 7.6 | 1.6 | 27.1 | 6.4 | 20.3 | 1.9 | 4.7 | 0 | — | |
| Aria-GUIParadigm=Trained GUI Action Models2025.10 | 11.3 | — | 16.2 | 0 | 23.7 | 2.1 | 7.6 | 1.6 | 27.1 | 6.4 | 20.3 | 1.9 | 4.7 | 0 | — | |
| CogAgent-18BProtocol=Supervised Fine-tuning, Examples=222M2025.07 | 7.7 | — | 14.9 | 0.7 | 9.6 | 0 | 7.1 | 3.1 | 22.2 | 1.8 | 13 | 0 | 5.6 | 0 | — | |
| ShowUI-2BProtocol=Supervised Fine-tuning, Examples=256K2025.07 | 7.7 | — | 16.9 | 1.4 | 9.1 | 0 | 2.5 | 0 | 13.2 | 7.3 | 15.3 | 7.5 | 10.3 | 2.2 | — | |
| ShowUI-2BParadigm=Trained GUI Action Models2025.10 | 7.7 | — | 16.9 | 1.4 | 9.1 | 0 | 2.5 | 0 | 13.2 | 7.3 | 15.3 | 7.5 | 10.3 | 2.2 | — | |
| CogAgent-18BParadigm=Trained GUI Action Models2025.10 | 7.7 | — | 14.9 | 0.7 | 9.6 | 0 | 7.1 | 3.1 | 22.2 | 1.8 | 13 | 0 | 5.6 | 0 | — | |
| OS-Atlas-4BParadigm=Trained GUI Action Models2025.10 | 3.7 | — | 7.1 | 0 | 3 | 1.4 | 2 | 0 | 9 | 5.5 | 5.1 | 3.8 | 5.6 | 0 | — | |
| MiniCPM-VParadigm=Generalist Models or Training-free Methods2025.10 | 3 | — | 7.1 | 0 | 2 | 0 | 4.1 | 1.6 | 8.3 | 0 | 2.8 | 3.8 | 3.7 | 1.1 | — | |
| Qwen2-VL-7BParadigm=Generalist Models or Training-free Methods2025.10 | 1.6 | — | 2.6 | 0 | 1.5 | 0 | 0.5 | 0 | 6.3 | 0 | 3.4 | 1.9 | 0.9 | 0 | — | |
| SeeClickParadigm=Trained GUI Action Models2025.10 | 1.1 | — | 0.6 | 0 | 1 | 0 | 2.5 | 0 | 3.5 | 0 | 1.1 | 0 | 2.8 | 0 | — | |
| GPT-4oParadigm=Generalist Models or Training-free Methods2025.10 | 0.8 | — | 1.3 | 0 | 1 | 0 | 2 | 0 | 2.1 | 0 | 1.1 | 0 | 0 | 0 | — | |
| Qwen-VL-7BParadigm=Generalist Models or Training-free Methods2025.10 | 0.1 | — | 0 | 0 | 0 | 0 | 0 | 0 | 0.7 | 0 | 0 | 0 | 0 | 0 | — | |
| CogAgent (18B)Platform=Multi-OS, Data=Real + Synthetic, # Images=40M2025.10 | — | 7.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Fara1.5-27BModel=Fara1.5-27B, Zoom procedure=with zoom2026.06 | — | 66.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Fara1.5-27BModel=Fara1.5-27B, Zoom procedure=without zoom2026.06 | — | 58.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Fara1.5-4BModel=Fara1.5-4B, Zoom procedure=with zoom2026.06 | — | 59.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Fara1.5-4BModel=Fara1.5-4B, Zoom procedure=without zoom2026.06 | — | 55.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Fara1.5-9BModel=Fara1.5-9B, Zoom procedure=with zoom2026.06 | — | 66.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Fara1.5-9BModel=Fara1.5-9B, Zoom procedure=without zoom2026.06 | — | 58 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Florence-2 (0.7B)Platform=Multi-OS, Data=Real + Synthetic, # Images=-2025.10 | — | 0.12 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUIrilla-See (0.7B)Platform=macOS, Data=Synthetic, # Images=4.2K2025.10 | — | 7.34 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUIrilla-See (3B)Platform=macOS, Data=Synthetic, # Images=4.2K2025.10 | — | 29.35 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GUIrilla-See (7B)Platform=macOS, Data=Synthetic, # Images=4.2K2025.10 | — | 35.36 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OS-Atlas (7B)Platform=Multi-OS, Data=Synthetic, # Images=2.2M2025.10 | — | 18.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen 2.5 VL (3B)Platform=Multi-OS, Data=Real + Synthetic, # Images=-2025.10 | — | 15.93 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen 2.5 VL (7B)Platform=Multi-OS, Data=Real + Synthetic, # Images=-2025.10 | — | 27.13 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ShowUI (2B)Platform=Multi-OS, Data=Real + Synthetic, # Images=256K2025.10 | — | 7.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UGround (7B)Platform=Web + Android, Data=Real, # Images=1.3M2025.10 | — | 16.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UI-TARS (2B)Platform=Multi-OS, Data=Real + Synthetic, # Images=∼20M (est.)2025.10 | — | 27.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UI-TARS (72B)Platform=Multi-OS, Data=Real + Synthetic, # Images=∼20M (est.)2025.10 | — | 38.1 | — | — | — | — | — | — | — | — | — | — | — | — | — |