Knowledge on ARC Challenge (ARC-C)
96.43ARC-C ScoreReasonAny
Evaluation results
| Method | Links | |
|---|---|---|
| ReasonAnyModel Family=Qwen2.5-32B2026.01 | 96.43 | |
| DAREModel Family=Qwen2.5-32B2026.01 | 96.27 | |
| TIESModel Family=Qwen2.5-32B2026.01 | 95.93 | |
| LEDModel Family=Qwen2.5-32B2026.01 | 95.88 | |
| Qwen2.5-32B-Instruct (Safety)Model Family=Qwen2.5-32B2026.01 | 95.59 | |
| Task ArithmeticModel Family=Qwen2.5-32B2026.01 | 95.59 | |
| FuseLLMModel Family=Qwen2.5-32B2026.01 | 95.59 | |
| LinearModel Family=Qwen2.5-32B2026.01 | 95.25 | |
| DeepSeek-R1-Distill-Qwen-32B (Reasoning)Model Family=Qwen2.5-32B2026.01 | 94.7 | |
| ReasonAnyModel Family=Qwen2.5-14B2026.01 | 92.52 | |
| LEDModel Family=Qwen2.5-14B2026.01 | 92.22 | |
| Qwen2.5-14B-Instruct (Safety)Model Family=Qwen2.5-14B2026.01 | 92.21 | |
| LinearModel Family=Qwen2.5-14B2026.01 | 91.97 | |
| DAREModel Family=Qwen2.5-14B2026.01 | 91.59 | |
| FuseLLMModel Family=Qwen2.5-14B2026.01 | 91.29 | |
| Task ArithmeticModel Family=Qwen2.5-14B2026.01 | 88.47 | |
| DeepSeek-R1-Distill-Qwen-14B (Reasoning)Model Family=Qwen2.5-14B2026.01 | 88.41 | |
| TIESModel Family=Qwen2.5-14B2026.01 | 85.87 | |
| Task ArithmeticMerging Protocol=Task Arithmetic2026.01 | 84.07 | |
| LEDMerging Protocol=LED2026.01 | 80.34 | |
| ReasonAnyMerging Protocol=ReasonAny2026.01 | 80.34 | |
| Qwen3-1.7B-ALLMEMModel Size=1.7B, Architecture=ALLMEM, Sliding window length=2048, Attention sinks=128, Few-shot setting=0-shot, Max tokens=327682026.02 | 74.4 | |
| Qwen3-1.7BModel Size=1.7B, Architecture=Standard, Few-shot setting=0-shot, Max tokens=327682026.02 | 74.2 | |
| ReasoningBase Model=Llama-3.1-8B2026.01 | 70.85 | |
| DAREMerging Protocol=DARE2026.01 | 64.41 | |
| TIESMerging Protocol=TIES2026.01 | 60 | |
| FuseLLMMerging Protocol=FuseLLM2026.01 | 55.25 | |
| Qwen3-0.6BModel Size=0.6B, Architecture=Standard, Few-shot setting=0-shot, Max tokens=327682026.02 | 55 | |
| Qwen3-0.6B-ALLMEMModel Size=0.6B, Architecture=ALLMEM, Sliding window length=2048, Attention sinks=128, Few-shot setting=0-shot, Max tokens=327682026.02 | 54.6 | |
| SafetyBase Model=Llama-3.1-8B2026.01 | 35.59 | |
| LinearMerging Protocol=Linear2026.01 | 30.85 |