@estebankiwiGPT-Image-2 has claimed the top spot on LMArena for image editing with an impressive 1510 Elo rating. It's 117 points ahead of second place and 123 points in front of Gemini 3 Pro, meaning OpenAI completely dominated the competition. If GPT-5.5 performs similarly in coding, Anthropic's advantage could vanish today, and people are testing it live right away.
View original postPhoto by @estebankiwi· Apr 23, 2026· GPT
The focus is on a leaderboard displaying various image editing models. The top position is highlighted with a red box, showing the model "gpt-image-2 (medium)" with a score of 1510. The background is dark, and the text is white, creating a high-contrast, clean aesthetic. The models are listed in descending order of their scores, which are displayed next to each model name. The interface has a modern, technical look, with icons representing each model and a "View all" option at the bottom.
@estebankiwiGPT-Image-2 has claimed the top spot on LMArena for image editing with an impressive 1510 Elo rating. It's 117 points ahead of second place and 123 points in front of Gemini 3 Pro, meaning OpenAI completely dominated the competition. If GPT-5.5 performs similarly in coding, Anthropic's advantage could vanish today, and people are testing it live right away.
View original postPhoto by @estebankiwi· Apr 23, 2026· GPT
The focus is on a leaderboard displaying various image editing models. The top position is highlighted with a red box, showing the model "gpt-image-2 (medium)" with a score of 1510. The background is dark, and the text is white, creating a high-contrast, clean aesthetic. The models are listed in descending order of their scores, which are displayed next to each model name. The interface has a modern, technical look, with icons representing each model and a "View all" option at the bottom.
