List your company
ResearchFormerly WizWand

Reasoning Image Editing on RISEBench 1.0 (test)

34.1Temporal ScoreGPT-4o
-1.3647.84317.0526.257Dec 5, 2025Dec 9, 2025Dec 13, 2025Dec 17, 2025Dec 21, 2025Dec 25, 2025Dec 29, 2025
Updated 8mo ago

Evaluation results

MethodLinks
GPT-4oAccess Type=Closed-source, Chain-of-Thought=false2025.1234.132.23710.628.9
GPT-Image-1Model Category=Proprietary Models2025.1234.132.23710.628.9
Gemini-2.5-Flash-ImageModel Category=Proprietary Models2025.1225.947.83718.832.8
EditThinkerBackbone=Qwen-Image-Edit, Framework=Think-while-Edit, Expert Model=EditThinker-Expert-GPT4.1, Protocol=Oracle Upper Bound2025.1225.932.2409.427.5
EditThinkerBackbone=OmniGen2, Framework=Think-while-Edit, Expert Model=EditThinker-Expert-GPT4.1, Protocol=Oracle Upper Bound2025.1217.68.982.49.2
EditThinkerBackbone=FLUX.1-Kontext-Dev, Framework=Think-while-Edit, Expert Model=EditThinker-Expert-GPT4.1, Protocol=Oracle Upper Bound2025.1216.524.4335.920.6
ThinkGenAccess Type=Open-source, Chain-of-Thought=true2025.1216.417.7161.113
Seedream-4.0Model Category=Proprietary Models2025.1212.912.2117.110.8
EditThinkerBackbone=FLUX.1-Kontext-Dev, Framework=Think-while-Edit, Expert Model=EditThinker-8B2025.1211.817.8207.114.4
EditThinkerBackbone=Qwen-Image-Edit, Framework=Think-while-Edit, Expert Model=EditThinker-8B2025.1210.823.3278.217.8
Gemini-2.0Access Type=Closed-source, Chain-of-Thought=false2025.128.215.5234.713.3
BAGELAccess Type=Open-source, Chain-of-Thought=true2025.125.917.8211.211.9
BAGELModel Category=Open-source Models, Chain-of-Thought=true2025.125.917.8211.211.9
Qwen-Image-EditModel Category=Open-source Models2025.124.710172.48.9
EditThinkerBackbone=OmniGen2, Framework=Think-while-Edit, Expert Model=EditThinker-8B2025.124.77.853.53.4
Qwen-Image-EditFramework=Think-while-Edit, Testing Protocol=Official model checkpoint test2025.124.710172.48.9
BAGELAccess Type=Open-source, Chain-of-Thought=false2025.123.54.495.95.8
ThinkGenAccess Type=Open-source, Chain-of-Thought=false2025.123.52.271.13.6
BAGELModel Category=Open-source Models2025.122.45.6141.26.1
OmniGen2Framework=Think-while-Edit, Testing Protocol=Official model checkpoint test2025.122.41.171.23.1
FLUX.1-Kontext-DevModel Category=Open-source Models2025.122.35.5131.25.8
FLUX.1-Kontext-DevFramework=Think-while-Edit, Testing Protocol=Official model checkpoint test2025.122.35.5131.25.8
OmniGenAccess Type=Open-source, Chain-of-Thought=false2025.121.2101.20.8
EMU2Access Type=Open-source, Chain-of-Thought=false2025.121.21.1000.5
Ovis-U1Model Category=Open-source Models2025.121.23.342.42.8
Step1X-EditAccess Type=Open-source, Chain-of-Thought=false2025.1202.223.51.9
HiDream-EditAccess Type=Open-source, Chain-of-Thought=false2025.1200000
FLUX-CannyAccess Type=Open-source, Chain-of-Thought=false2025.1200000
OmniGen2Access Type=Open-source, Chain-of-Thought=false2025.1202.272.33
Step1X-EditModel Category=Open-source Models2025.1202.223.51.9