Multi-task Language Evaluation on SEA-HELM
83Instruction Following (SEA-IFEval)SiamGPT-32B
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| SiamGPT-32BNumber of Parameters=32B, Backbone Architecture=Qwen32025.12 | 83 | 75.81 | 42.06 | 67.95 | 68.59 | 44.19 | 63.6 | |
| Typhoon 2.5Number of Parameters=30B, Backbone Architecture=Qwen32025.12 | 79 | 76.16 | 56.7 | 65.56 | 55.54 | 29.68 | 60.44 | |
| OTG-R1Number of Parameters=32B, Backbone Architecture=DeepSeek-R12025.12 | 54 | 59.69 | 54.31 | 59.89 | 65.38 | 41.42 | 55.78 |