List your company

VLM Pairwise Preference on ArcBench

100Win RateArcDeck
23.76843.55963.3583.141Apr 13, 2026

Evaluation results

MethodLinks
ArcDeckGeneration Model=GPT-5, Evaluation Model=Qwen3, Baseline=PPTAgent2026.04100
ArcDeckGeneration Model=GPT-5, Evaluation Model=GPT-5, Baseline=PPTAgent2026.04100
ArcDeckGeneration Model=Qwen3, Evaluation Model=Qwen3, Baseline=PPTAgent2026.04100
ArcDeckGeneration Model=GPT-4o, Evaluation Model=Qwen3, Baseline=PPTAgent2026.0499
ArcDeckGeneration Model=Qwen3, Evaluation Model=GPT-5, Baseline=PPTAgent2026.0497.9
ArcDeckGeneration Model=GPT-5, Evaluation Model=GPT-5, Baseline=P2Poster2026.0497.6
ArcDeckGeneration Model=GPT-4o, Evaluation Model=GPT-5, Baseline=PPTAgent2026.0495
ArcDeckGeneration Model=Qwen3, Evaluation Model=Qwen3, Baseline=P2Poster2026.0492.8
ArcDeckGeneration Model=GPT-4o, Evaluation Model=GPT-5, Baseline=HTML2026.0492
ArcDeckGeneration Model=Qwen3, Evaluation Model=GPT-5, Baseline=P2Poster2026.0490.7
ArcDeckGeneration Model=GPT-5, Evaluation Model=GPT-5, Baseline=HTML2026.0484.7
ArcDeckGeneration Model=Qwen3, Evaluation Model=GPT-5, Baseline=HTML2026.0484.5
ArcDeckGeneration Model=GPT-4o, Evaluation Model=GPT-5, Baseline=P2Poster2026.0484
ArcDeckGeneration Model=GPT-5, Evaluation Model=Qwen3, Baseline=HTML2026.0482.4
ArcDeckGeneration Model=GPT-4o, Evaluation Model=Qwen3, Baseline=SlideGen2026.0481
ArcDeckGeneration Model=GPT-4o, Evaluation Model=Qwen3, Baseline=HTML2026.0479
ArcDeckGeneration Model=GPT-4o, Evaluation Model=Qwen3, Baseline=P2Poster2026.0478
ArcDeckGeneration Model=Qwen3, Evaluation Model=Qwen3, Baseline=HTML2026.0476.3
ArcDeckGeneration Model=GPT-4o, Evaluation Model=GPT-5, Baseline=SlideGen2026.0468
ArcDeckGeneration Model=GPT-5, Evaluation Model=Qwen3, Baseline=P2Poster2026.0462.4
ArcDeckGeneration Model=Qwen3, Evaluation Model=Qwen3, Baseline=SlideGen2026.0457.7
ArcDeckGeneration Model=GPT-5, Evaluation Model=Qwen3, Baseline=SlideGen2026.0455.3
ArcDeckGeneration Model=GPT-5, Evaluation Model=GPT-5, Baseline=SlideGen2026.0450.6
ArcDeckGeneration Model=Qwen3, Evaluation Model=GPT-5, Baseline=SlideGen2026.0448.5
ArcDeckGeneration Model=GPT-5, Evaluation Model=GPT-52026.0448.1
ArcDeckGeneration Model=Qwen3, Evaluation Model=Qwen32026.0447.3
SlideGenGeneration Model=GPT-5, Evaluation Model=Qwen32026.0446.8
ArcDeckGeneration Model=Qwen3, Evaluation Model=GPT-52026.0446.7
SlideGenGeneration Model=Qwen3, Evaluation Model=Qwen32026.0446
SlideGenGeneration Model=GPT-5, Evaluation Model=GPT-52026.0445.8
SlideGenGeneration Model=Qwen3, Evaluation Model=GPT-52026.0445.8
PPTAgentGeneration Model=GPT-5, Evaluation Model=Qwen32026.0445.6
PPTAgentGeneration Model=Qwen3, Evaluation Model=Qwen32026.0445.4
HTMLGeneration Model=Qwen3, Evaluation Model=Qwen32026.0442.6
P2PosterGeneration Model=Qwen3, Evaluation Model=GPT-52026.0440
PPTAgentGeneration Model=GPT-5, Evaluation Model=GPT-52026.0440
HTMLGeneration Model=Qwen3, Evaluation Model=GPT-52026.0440
ArcDeckGeneration Model=GPT-4o, Evaluation Model=GPT-52026.0440
P2PosterGeneration Model=Qwen3, Evaluation Model=Qwen32026.0437.7
PPTAgentGeneration Model=Qwen3, Evaluation Model=GPT-52026.0436.7
ArcDeckGeneration Model=GPT-4o, Evaluation Model=Qwen32026.0436.6
HTMLGeneration Model=GPT-5, Evaluation Model=Qwen32026.0435.9
ArcDeckGeneration Model=GPT-5, Evaluation Model=Qwen32026.0435.7
PPTAgentGeneration Model=GPT-4o, Evaluation Model=Qwen32026.0435.4
P2PosterGeneration Model=GPT-5, Evaluation Model=Qwen32026.0435.1
P2PosterGeneration Model=GPT-5, Evaluation Model=GPT-52026.0433.3
SlideGenGeneration Model=GPT-4o, Evaluation Model=GPT-52026.0433.3
HTMLGeneration Model=GPT-5, Evaluation Model=GPT-52026.0433.3
HTMLGeneration Model=GPT-4o, Evaluation Model=GPT-52026.0433.3
SlideGenGeneration Model=GPT-4o, Evaluation Model=Qwen32026.0432.1
HTMLGeneration Model=GPT-4o, Evaluation Model=Qwen32026.0431.3
PPTAgentGeneration Model=GPT-4o, Evaluation Model=GPT-52026.0430
P2PosterGeneration Model=GPT-4o, Evaluation Model=Qwen32026.0427.5
P2PosterGeneration Model=GPT-4o, Evaluation Model=GPT-52026.0426.7