Mathematical Reasoning on AIME24 (TPF, AVG@8)
5.07TPFTDAR-8B-thinking + BACD
Evaluation results
| Method | Links | ||
|---|---|---|---|
| TDAR-8B-thinking + BACDModel Type=Block Diffusion, Block Size (B)=16, Decoding=BACD2026.02 | 5.07 | 36.3 | |
| TDAR-8B-thinkingModel Type=Block Diffusion, Block Size (B)=162026.02 | 4.47 | 34.6 | |
| LLaDAModel Type=Masked Diffusion2026.02 | 3.44 | 6.7 | |
| LLaDA-1.5Model Type=Masked Diffusion2026.02 | 3.34 | 0 | |
| TDAR-8B-thinking + BACD + TCCFModel Type=Block Diffusion, Block Size (B)=16, Decoding=BACD + TCCF2026.02 | 3.04 | 42.9 | |
| SDAR-8B-ChatModel Type=Block Diffusion, Block Size (B)=42026.02 | 2.96 | 5 | |
| LLaDA-MoEModel Type=Masked Diffusion2026.02 | 2.89 | 3.3 | |
| Fast-dLLM-v2Model Type=Block Diffusion, Block Size (B)=322026.02 | 2.58 | 0 | |
| TraDo-8B-InstructModel Type=Block Diffusion, Block Size (B)=42026.02 | 2.13 | 13.3 | |
| DiRL-8B-InstructModel Type=Block Diffusion, Block Size (B)=42026.02 | 1.96 | 18.8 | |
| TraDo-8B-Thinking + BACDModel Type=Block Diffusion, Block Size (B)=4, Decoding=BACD2026.02 | 1.44 | 32.9 | |
| TraDo-8B-Thinking + BACD + TCCFModel Type=Block Diffusion, Block Size (B)=4, Decoding=BACD + TCCF2026.02 | 1.36 | 35.8 | |
| TraDo-8B-ThinkingModel Type=Block Diffusion, Block Size (B)=42026.02 | 1.35 | 31.3 | |
| Qwen3-8B-Thinking†Model Type=Autoregressive2026.02 | 1 | 63.3 |