Instruction Following on ReasonIF synthesized v1.0
96.3IFSFLEx
Evaluation Results
| Method | Links | |
|---|---|---|
| FLExBase Model=Qwen-32B, Model Family=Qwen-2.5 Instruct2026.01 | 96.3 | |
| FLExBase Model=Qwen-14B, Model Family=Qwen-2.5 Instruct2026.01 | 95 | |
| FLExBase Model=Qwen-72B, Model Family=Qwen-2.5 Instruct2026.01 | 93.7 | |
| CoTBase Model=Qwen-72B, Model Family=Qwen-2.5 Instruct2026.01 | 87 | |
| SRBase Model=Qwen-72B, Model Family=Qwen-2.5 Instruct2026.01 | 85.3 | |
| RAGBase Model=Qwen-72B, Model Family=Qwen-2.5 Instruct2026.01 | 83.7 | |
| SCBase Model=Qwen-72B, Model Family=Qwen-2.5 Instruct2026.01 | 83 | |
| RAGBase Model=Qwen-32B, Model Family=Qwen-2.5 Instruct2026.01 | 79.7 | |
| SCBase Model=Qwen-32B, Model Family=Qwen-2.5 Instruct2026.01 | 79 | |
| CoTBase Model=Qwen-32B, Model Family=Qwen-2.5 Instruct2026.01 | 78.3 | |
| RAGBase Model=Gemma-27B, Model Family=Gemma-3 Instruct2026.01 | 77.3 | |
| FLExBase Model=Gemma-27B, Model Family=Gemma-3 Instruct2026.01 | 76.7 | |
| RAGBase Model=Qwen-14B, Model Family=Qwen-2.5 Instruct2026.01 | 75.3 | |
| CoTBase Model=Gemma-27B, Model Family=Gemma-3 Instruct2026.01 | 75 | |
| SRBase Model=Qwen-32B, Model Family=Qwen-2.5 Instruct2026.01 | 75 | |
| FLExBase Model=Gemma-12B, Model Family=Gemma-3 Instruct2026.01 | 74 | |
| SCBase Model=Gemma-27B, Model Family=Gemma-3 Instruct2026.01 | 73.7 | |
| SCBase Model=Qwen-14B, Model Family=Qwen-2.5 Instruct2026.01 | 72.3 | |
| CoTBase Model=Qwen-14B, Model Family=Qwen-2.5 Instruct2026.01 | 71.3 | |
| RAGBase Model=Gemma-12B, Model Family=Gemma-3 Instruct2026.01 | 71 | |
| SRBase Model=Qwen-14B, Model Family=Qwen-2.5 Instruct2026.01 | 70 | |
| SRBase Model=Gemma-27B, Model Family=Gemma-3 Instruct2026.01 | 68.7 | |
| FLExBase Model=Qwen-7B, Model Family=Qwen-2.5 Instruct2026.01 | 67.7 | |
| RAGBase Model=Qwen-7B, Model Family=Qwen-2.5 Instruct2026.01 | 66.7 | |
| FLExBase Model=Gemma-4B, Model Family=Gemma-3 Instruct2026.01 | 66 | |
| CoTBase Model=Gemma-12B, Model Family=Gemma-3 Instruct2026.01 | 64.3 | |
| SCBase Model=Gemma-12B, Model Family=Gemma-3 Instruct2026.01 | 63.3 | |
| RAGBase Model=Gemma-4B, Model Family=Gemma-3 Instruct2026.01 | 58 | |
| SRBase Model=Gemma-12B, Model Family=Gemma-3 Instruct2026.01 | 58 | |
| SRBase Model=Qwen-7B, Model Family=Qwen-2.5 Instruct2026.01 | 56.7 | |
| FLExBase Model=Gemma-1B, Model Family=Gemma-3 Instruct2026.01 | 50 | |
| CoTBase Model=Gemma-4B, Model Family=Gemma-3 Instruct2026.01 | 50 | |
| RAGBase Model=Qwen-1.5B, Model Family=Qwen-2.5 Instruct2026.01 | 48 | |
| SRBase Model=Gemma-4B, Model Family=Gemma-3 Instruct2026.01 | 47.3 | |
| SCBase Model=Qwen-7B, Model Family=Qwen-2.5 Instruct2026.01 | 47.3 | |
| SCBase Model=Gemma-4B, Model Family=Gemma-3 Instruct2026.01 | 46.7 | |
| RAGBase Model=Gemma-1B, Model Family=Gemma-3 Instruct2026.01 | 46.3 | |
| CoTBase Model=Qwen-7B, Model Family=Qwen-2.5 Instruct2026.01 | 46.3 | |
| RAGBase Model=Qwen-0.5B, Model Family=Qwen-2.5 Instruct2026.01 | 45 | |
| SRBase Model=Qwen-3B, Model Family=Qwen-2.5 Instruct2026.01 | 44.7 | |
| FLExBase Model=Qwen-3B, Model Family=Qwen-2.5 Instruct2026.01 | 42.3 | |
| SCBase Model=Qwen-3B, Model Family=Qwen-2.5 Instruct2026.01 | 39.3 | |
| RAGBase Model=Qwen-3B, Model Family=Qwen-2.5 Instruct2026.01 | 38.7 | |
| SRBase Model=Gemma-1B, Model Family=Gemma-3 Instruct2026.01 | 38 | |
| SCBase Model=Gemma-1B, Model Family=Gemma-3 Instruct2026.01 | 36.3 | |
| SRBase Model=Qwen-1.5B, Model Family=Qwen-2.5 Instruct2026.01 | 36.3 | |
| CoTBase Model=Qwen-3B, Model Family=Qwen-2.5 Instruct2026.01 | 36.3 | |
| SRBase Model=Qwen-0.5B, Model Family=Qwen-2.5 Instruct2026.01 | 35.7 | |
| CoTBase Model=Gemma-1B, Model Family=Gemma-3 Instruct2026.01 | 35.3 | |
| FLExBase Model=Qwen-0.5B, Model Family=Qwen-2.5 Instruct2026.01 | 35.3 | |
| FLExBase Model=Qwen-1.5B, Model Family=Qwen-2.5 Instruct2026.01 | 35 | |
| SCBase Model=Qwen-1.5B, Model Family=Qwen-2.5 Instruct2026.01 | 31.3 | |
| CoTBase Model=Qwen-1.5B, Model Family=Qwen-2.5 Instruct2026.01 | 30.3 | |
| SCBase Model=Qwen-0.5B, Model Family=Qwen-2.5 Instruct2026.01 | 29 | |
| CoTBase Model=Qwen-0.5B, Model Family=Qwen-2.5 Instruct2026.01 | 28.3 |