Reasoning on CounterBench
0.0359Error RateFLEx
Evaluation Results
| Method | Links | |
|---|---|---|
| FLExBackbone Model=Gemma-1B2026.01 | 0.0359 | |
| FLExBackbone Model=Qwen-1.5B2026.01 | 0.14 | |
| FLExBackbone Model=Qwen-3B2026.01 | 0.1535 | |
| FLExBackbone Model=Gemma-27B2026.01 | 0.1958 | |
| FLExBackbone Model=Qwen-32B2026.01 | 0.2414 | |
| FLExBackbone Model=Qwen-14B2026.01 | 0.2538 | |
| FLExBackbone Model=Gemma-4B2026.01 | 0.2678 | |
| FLExBackbone Model=Qwen-7B2026.01 | 0.2691 | |
| FLExBackbone Model=Qwen-0.5B2026.01 | 0.2706 | |
| FLExBackbone Model=Qwen-72B2026.01 | 0.3547 | |
| FLExBackbone Model=Gemma-12B2026.01 | 0.5176 |