{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/1a70a45a69d24007b624c537182295d2\" frameborder=\"0\" width=\"1912\" height=\"1434\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1434,"width":1912,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1434,"thumbnail_width":1912,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/1a70a45a69d24007b624c537182295d2-e41ef246e340d81e.gif","duration":335.736,"title":"Automating GPT Image 2 Infographics from Transcripts","description":"This Loom discusses how to generate GPT Image 2.0 style infographic outputs from coaching and course transcripts, aiming to automate what the creator previously did in another tool. The author asks whether the model setting matters when an action already specifies Image 2.0, considering they want to use GPT-5 required on ChatGPT. They also question whether “thinking” shown in the action is the same as the image generator’s prompt reasoning, and whether prompt complexity or action configuration affects triggering and results. The Loom references a prior working setup where reasoning steps and a widescreen format were enabled, and it includes an example of a transcript test to evaluate behavior."}