List your company

Roadmap generation on RoadMapper Combined English & Chinese

79.61Overall AverageRoadMapper
59.860464.987770.11575.2423Apr 30, 2026

Evaluation results

MethodLinks
RoadMapperBase Model=DeepSeek-V3.2, Training=DPO training on Qwen3-32B2026.0479.61
RoadMapperBase Model=Qwen3-235B-A22B, Training=DPO training on Qwen3-32B2026.0478.71
DyLANBase Model=DeepSeek-V3.22026.0477.36
RoadMapperBase Model=Gemini 3 Flash Preview, Training=DPO training on Qwen3-32B2026.0477.29
ReConcileBase Model=DeepSeek-V3.22026.0476.74
DyLANBase Model=Qwen3-235B-A22B2026.0475.89
ReConcileBase Model=Qwen3-235B-A22B2026.0475.85
RoadMapperBase Model=gpt-oss-20b, Training=DPO training on Qwen3-32B2026.0475.26
RoadMapperBase Model=Qwen3-14B, Training=DPO training on Qwen3-32B2026.0475.08
ReConcileBase Model=Gemini 3 Flash Preview2026.0473.67
CoTBase Model=Qwen3-235B-A22B2026.0473.47
DyLANBase Model=Gemini 3 Flash Preview2026.0473.42
Best-of-NBase Model=Qwen3-235B-A22B2026.0473.4
Direct PromptingBase Model=Qwen3-235B-A22B2026.0472.91
CoTBase Model=DeepSeek-V3.22026.0472.75
Best-of-NBase Model=DeepSeek-V3.22026.0472.58
RoadMapperBase Model=Claude 3 Haiku, Training=DPO training on Qwen3-32B2026.0472.55
RoadMapperBase Model=Mistral Small 3.2, Training=DPO training on Qwen3-32B2026.0472.33
Direct PromptingBase Model=DeepSeek-V3.22026.0472.12
RoadMapperBase Model=Llama 4 Maverick, Training=DPO training on Qwen3-32B2026.0470.85
Best-of-NBase Model=Qwen3-14B2026.0470.53
RoadMapperBase Model=Llama 3.3 70B, Training=DPO training on Qwen3-32B2026.0470.45
Best-of-NBase Model=gpt-oss-20b2026.0470.25
Direct PromptingBase Model=Qwen3-14B2026.0470.21
CoTBase Model=Gemini 3 Flash Preview2026.0469.87
Direct PromptingBase Model=gpt-oss-20b2026.0469.8
Best-of-NBase Model=Gemini 3 Flash Preview2026.0469.55
Direct PromptingBase Model=Gemini 3 Flash Preview2026.0469.2
RoadMapperBase Model=Llama 3.1 8B, Training=DPO training on Qwen3-32B2026.0467.82
RoadMapperBase Model=GPT-4o mini, Training=DPO training on Qwen3-32B2026.0467.18
Best-of-NBase Model=Claude 3 Haiku2026.0466.92
Best-of-NBase Model=Mistral Small 3.22026.0466.84
Direct PromptingBase Model=Mistral Small 3.22026.0466.65
DyLANBase Model=Llama 3.3 70B2026.0466.46
Direct PromptingBase Model=Claude 3 Haiku2026.0466.43
ReConcileBase Model=Llama 3.3 70B2026.0466.25
Best-of-NBase Model=Llama 4 Maverick2026.0463.51
Direct PromptingBase Model=Llama 4 Maverick2026.0463.38
Best-of-NBase Model=GPT-4o mini2026.0463.36
Direct PromptingBase Model=GPT-4o mini2026.0462.95
CoTBase Model=Llama 3.3 70B2026.0462.88
Best-of-NBase Model=Llama 3.3 70B2026.0462.74
Direct PromptingBase Model=Llama 3.3 70B2026.0462.1
Best-of-NBase Model=Llama 3.1 8B2026.0461.36
Direct PromptingBase Model=Llama 3.1 8B2026.0460.62