Reasoning-informed Image Editing on RISE-Bench
57.6Temporal ScoreGPT-Image-1.5
Evaluation Results
| Method | Links | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| GPT-Image-1.5Model Type=Closed-source proprietary2026.05 | 57.6 | 62.2 | 62 | 21.2 | 51.4 | — | — | — | — | 69.7 | 92.5 | 94.9 | |
| GPT-Image-1.5Chain-of-Thought (CoT)=false2026.05 | 54.1 | 60 | 62 | 21.2 | 50 | — | — | — | — | 69.7 | 92.5 | 94.9 | |
| DDA-Thinker-32BModel Category=Community Models, Base Editor=Qwen-Image-Edit-2511, Thinker Model=DDA-Thinker, Thinker Size=32B2026.04 | 45.9 | 50 | 47 | 15.3 | — | 68.4 | 84.4 | 92.1 | 40 | — | — | — | |
| GPT-Image-2Model Type=Closed-source proprietary2026.05 | 45.9 | 66.7 | 50 | 34.1 | 49.4 | — | — | — | — | 73.8 | 89.3 | 94.9 | |
| GPT-Image-2Source Type=Closed-Source2026.07 | 45.9 | 66.7 | 50 | 34.1 | 49.4 | — | — | — | — | — | — | — | |
| Gemini-3-pro-imageModel Type=Closed-source proprietary2026.05 | 43.5 | 63.3 | 48 | 37.6 | 48.3 | — | — | — | — | 77 | 85.5 | 94.4 | |
| Nano Banana ProSource Type=Closed-Source2026.07 | 43.5 | 63.3 | 48 | 37.6 | 48.3 | — | — | — | — | — | — | — | |
| Gemini-3-Pro-Image (NanoBanana-Pro)Model Category=Proprietary Models2026.04 | 41.2 | 61.1 | 48 | 37.6 | — | 77 | 85.5 | 94.4 | 47.2 | — | — | — | |
| Nano-Banana-ProChain-of-Thought (CoT)=false2026.05 | 41.2 | 61.1 | 48 | 37.6 | 47.2 | — | — | — | — | 77 | 85.5 | 94.4 | |
| Critic-v1Model Type=Open-source editor2026.05 | 37.6 | 46.7 | 42 | 8.2 | 34.2 | — | — | — | — | 59.1 | 85.4 | 92.2 | |
| GPT-Image-1Model Type=Closed-source proprietary2026.05 | 36.5 | 34.4 | 37 | 10.6 | 30 | — | — | — | — | 62.8 | 80.2 | 94.9 | |
| GPT-Image-1Source Type=Closed-Source2026.07 | 36.5 | 34.4 | 37 | 10.6 | 30 | — | — | — | — | — | — | — | |
| Critic-v2Model Type=Open-source editor2026.05 | 35.3 | 47.8 | 50 | 15.3 | 37.8 | — | — | — | — | 63.7 | 88.5 | 90.7 | |
| GPT-Image-1Model Category=Proprietary Models2026.04 | 34.1 | 32.2 | 37 | 10.6 | — | 62.8 | 80.2 | 94.9 | 28.9 | — | — | — | |
| DDA-Thinker-32BModel Category=Community Models, Base Editor=Qwen-Image-Edit-2509, Thinker Model=DDA-Thinker, Thinker Size=32B2026.04 | 34.1 | 36.7 | 47 | 15.1 | — | 67.2 | 79.2 | 89.5 | 33.9 | — | — | — | |
| GPT-Image-1Chain-of-Thought (CoT)=false2026.05 | 34.1 | 32.2 | 37 | 10.6 | 28.9 | — | — | — | — | 62.8 | 80.2 | 94.9 | |
| Unified ThinkerModel Category=Community Models2026.04 | 32.9 | 30 | 41 | 9.4 | — | 61.9 | 76.2 | 90.5 | 28.9 | — | — | — | |
| DDA-Thinker-8BModel Category=Community Models, Base Editor=Qwen-Image-Edit-2511, Thinker Model=DDA-Thinker, Thinker Size=8B2026.04 | 31.8 | 41.1 | 40 | 12.9 | — | 66.3 | 79.5 | 91.5 | 31.9 | — | — | — | |
| SenseNova-U1-SFT# Params=8B, Chain-of-Thought (CoT)=true2026.05 | 31.8 | 33.3 | 27 | 15.3 | 26.9 | — | — | — | — | 60.8 | 86.6 | 88.2 | |
| Gemini-2.5-Flash-ImageModel Type=Closed-source proprietary2026.05 | 29.4 | 48.9 | 37 | 18.8 | 33.9 | — | — | — | — | 61.2 | 86 | 91.3 | |
| discipline-informed reasoning-generation modelSource Type=Open-Source2026.07 | 28.2 | 25.6 | 40 | 9.4 | 26.4 | — | — | — | — | — | — | — | |
| Gemini-2.5-Flash-Image (NanoBanana)Model Category=Proprietary Models2026.04 | 25.9 | 47.8 | 37 | 18.8 | — | 61.2 | 86 | 91.3 | 32.8 | — | — | — | |
| DDA-Thinker-8BModel Category=Community Models, Base Editor=Qwen-Image-Edit-2509, Thinker Model=DDA-Thinker, Thinker Size=8B2026.04 | 25.9 | 40 | 42 | 10.6 | — | 60.8 | 78.5 | 88.4 | 30.3 | — | — | — | |
| Nano-BananaChain-of-Thought (CoT)=false2026.05 | 25.9 | 47.8 | 37 | 18.8 | 32.8 | — | — | — | — | 61.2 | 86 | 91.3 | |
| SenseNova-U1-SFT# Params=8BA3B, Chain-of-Thought (CoT)=false2026.05 | 25.9 | 41.1 | 26 | 7.1 | 25.3 | — | — | — | — | 57.4 | 82.6 | 85.4 | |
| GPT-Image-1-miniModel Category=Proprietary Models2026.04 | 24.7 | 28.9 | 33 | 9.4 | — | 54.1 | 71.5 | 93.7 | 24.4 | — | — | — | |
| GPT-Image-1-miniChain-of-Thought (CoT)=false2026.05 | 24.7 | 28.9 | 33 | 9.4 | 24.4 | — | — | — | — | 54.1 | 71.5 | 93.7 | |
| SenseNova-U1-SFT# Params=8BA3B, Chain-of-Thought (CoT)=true2026.05 | 24.7 | 46.7 | 28 | 20 | 30 | — | — | — | — | 63.2 | 84.1 | 87.4 | |
| SenseNova-U1-SFT# Params=8B, Chain-of-Thought (CoT)=false2026.05 | 22.4 | 33.3 | 27 | 11.8 | 23.9 | — | — | — | — | 58.2 | 84.1 | 82.4 | |
| Qwen-Image-Edit-2511Model Category=Community Models, Base Editor=Qwen-Image-Edit-25112026.04 | 21.2 | 18.9 | 31 | 4.7 | — | 49.9 | 71 | 91.5 | 19.4 | — | — | — | |
| Qwen-Image-Edit-2511# Params=20B, Chain-of-Thought (CoT)=false2026.05 | 21.2 | 18.9 | 31 | 4.7 | 19.4 | — | — | — | — | 49.9 | 71 | 91.5 | |
| Qwen-Image-Edit-2511Model Type=Open-source editor2026.05 | 21.2 | 18.9 | 31 | 4.7 | 19.4 | — | — | — | — | 49.9 | 71 | 91.5 | |
| Qwen-Image-Edit-2511Source Type=Open-Source2026.07 | 21.2 | 18.9 | 31 | 4.7 | 19.4 | — | — | — | — | — | — | — | |
| ThinkRL-EditBase Model=Qwen-Edit2026.01 | 18.8 | 37.5 | 25 | 37.5 | 29.7 | 61.7 | 81.64 | 62.5 | — | — | — | — | |
| ThinkRL-EditModel Category=Community Models2026.04 | 18.8 | 37.5 | 25 | 37.5 | — | 61.7 | 81.6 | 62.5 | 29.7 | — | — | — | |
| Seedream-4.0Source Type=Closed-Source2026.07 | 17.6 | 13.3 | 11 | 7.1 | 12.2 | — | — | — | — | — | — | — | |
| ThinkGenModel Category=Community Models2026.04 | 16.4 | 17.7 | 16 | 1.1 | — | — | — | — | 13 | — | — | — | |
| DeepGenModel Category=Community Models2026.04 | 15.3 | 18.9 | 14 | 4.7 | — | — | — | — | 13.3 | — | — | — | |
| Seedream 4.0Chain-of-Thought (CoT)=false2026.05 | 12.9 | 12.2 | 11 | 7.1 | 10.8 | — | — | — | — | 58.9 | 67.4 | 91.2 | |
| BAGELSource Type=Open-Source, thinking=true2026.07 | 11.8 | 27.8 | 21 | 1.2 | 15.8 | — | — | — | — | — | — | — | |
| EditThinkerModel Category=Community Models2026.04 | 10.8 | 23.3 | 27 | 8.2 | — | — | — | — | 17.8 | — | — | — | |
| Gemini-2.0-Flash-preChain-of-Thought (CoT)=false2026.05 | 10.6 | 13.3 | 11 | 2.3 | 9.4 | — | — | — | — | 49.9 | 68.4 | 84.9 | |
| UniCoT2026.01 | 8.2 | 18.9 | 20 | 1.2 | 12.5 | 48.3 | 76.2 | 83.8 | — | — | — | — | |
| Uni-CoTModel Category=Community Models2026.04 | 8.2 | 18.9 | 20 | 1.2 | — | — | — | — | 12.5 | — | — | — | |
| Gemini-2.0-Flash-expChain-of-Thought (CoT)=false2026.05 | 8.2 | 15.5 | 23 | 4.7 | 13.3 | — | — | — | — | 48.9 | 68.2 | 82.7 | |
| Bagel-Think2026.01 | 5.9 | 17.8 | 21 | 1.2 | 11.9 | 45.9 | 73.8 | 80.1 | — | — | — | — | |
| BAGEL (w/ CoT)Model Category=Community Models2026.04 | 5.9 | 17.8 | 21 | 1.2 | — | 45.9 | 73.8 | 80.1 | 11.9 | — | — | — | |
| BAGEL# Params=7B, Chain-of-Thought (CoT)=true2026.05 | 5.9 | 17.8 | 21 | 1.2 | 11.9 | — | — | — | — | 45.9 | 73.8 | 80.1 | |
| BAGEL (w/ CoT)Model Type=Open-source editor2026.05 | 5.9 | 17.8 | 21 | 1.2 | 11.9 | — | — | — | — | 45.9 | 73.8 | 80.1 | |
| Qwen-Edit2026.01 | 4.7 | 10 | 17 | 2.4 | 8.9 | 37.2 | 66.4 | 86.9 | — | — | — | — | |
| Qwen-Image-Edit-2509Model Category=Community Models, Base Editor=Qwen-Image-Edit-25092026.04 | 4.7 | 10 | 17 | 2.4 | — | 37.2 | 66.4 | 86.9 | 8.9 | — | — | — | |
| InternVL-U# Params=1.7B, Chain-of-Thought (CoT)=true2026.05 | 4.7 | 7.8 | 1.8 | 5.9 | 9.4 | — | — | — | — | 43.9 | 64.4 | 79.7 | |
| Qwen-Image-Edit-2509# Params=20B, Chain-of-Thought (CoT)=false2026.05 | 4.7 | 10 | 17 | 2.4 | 8.9 | — | — | — | — | 37.2 | 66.4 | 86.9 | |
| Qwen-Image-Edit-2509Model Type=Open-source editor2026.05 | 4.7 | 11.1 | 17 | 2.4 | 9.2 | — | — | — | — | 37.2 | 66.4 | 86.9 | |
| InternVL-U# Params=1.7B, Chain-of-Thought (CoT)=false2026.05 | 3.5 | 2.2 | 5 | 3.5 | 3.6 | — | — | — | — | 35.6 | 52.7 | 75.9 | |
| BAGELSource Type=Open-Source, thinking=false2026.07 | 3.5 | 4.4 | 14 | 0 | 5.8 | — | — | — | — | — | — | — | |
| Bagel2026.01 | 2.4 | 5.6 | 14 | 1.2 | 6.1 | 36.5 | 53.5 | 73 | — | — | — | — | |
| BAGEL# Params=7B, Chain-of-Thought (CoT)=false2026.05 | 2.4 | 5.6 | 14 | 1.2 | 6.1 | — | — | — | — | 36.5 | 53.5 | 73 | |
| Lumina-DiMOO# Params=8B, Chain-of-Thought (CoT)=false2026.05 | 2.4 | 1.1 | 4 | 1.2 | 2.2 | — | — | — | — | 34 | 50.7 | 72.3 | |
| Flux-Kontext2026.01 | 2.3 | 5.5 | 13 | 1.2 | 5.8 | 26 | 71.6 | 85.2 | — | — | — | — | |
| FLUX.1-Kontext-DevModel Category=Community Models2026.04 | 2.3 | 5.5 | 13 | 1.2 | — | 26 | 71.6 | 85.2 | 5.8 | — | — | — | |
| FLUX.1-Kontext-Dev# Params=12B, Chain-of-Thought (CoT)=false2026.05 | 2.3 | 5.5 | 13 | 1.2 | 5.8 | — | — | — | — | 26 | 71.6 | 85.2 | |
| FLUX.1-Kontext-DevModel Type=Open-source editor2026.05 | 2.3 | 5.5 | 13 | 1.2 | 5.8 | — | — | — | — | 26 | 71.6 | 85.2 | |
| OmniGen22026.01 | 1.2 | 1 | 0 | 1.2 | 0.8 | 22 | 32.6 | 55.3 | — | — | — | — | |
| EMU2Model Category=Community Models2026.04 | 1.2 | 1.1 | 0 | 0 | — | 22.6 | 38.2 | 78.3 | 0.5 | — | — | — | |
| OmniGenModel Category=Community Models2026.04 | 1.2 | 1 | 0 | 1.2 | — | 25.1 | 41.5 | 73.5 | 0.8 | — | — | — | |
| Ovis-U1Model Category=Community Models2026.04 | 1.2 | 3.3 | 4 | 2.4 | — | 33.9 | 52.7 | 72.9 | 2.8 | — | — | — | |
| Ovis-U1# Params=1.2B, Chain-of-Thought (CoT)=false2026.05 | 1.2 | 3.3 | 4 | 2.4 | 2.8 | — | — | — | — | 33.9 | 52.7 | 72.9 | |
| OmniGen# Params=3.8B, Chain-of-Thought (CoT)=false2026.05 | 1.2 | 1 | 0 | 1.2 | 0.8 | — | — | — | — | 22 | 32.6 | 55.3 | |
| Emu2# Params=37B, Chain-of-Thought (CoT)=false2026.05 | 1.2 | 1.1 | 0 | 0 | 0.5 | — | — | — | — | 22.6 | 38.2 | 78.3 | |
| Step1X-EditModel Category=Community Models2026.04 | 0 | 2.2 | 2 | 3.5 | — | 30.3 | 12.6 | 74.9 | 1.9 | — | — | — | |
| Step1X-Edit# Params=12B, Chain-of-Thought (CoT)=false2026.05 | 0 | 2.2 | 2 | 3.5 | 1.9 | — | — | — | — | 25.1 | 41.5 | 73.5 |