One-line prompt test shows GPT Image 2.5 producing near-indistinguishable iPhone-style photos
Scobleizer · x · 2026-09-09
User blueemi99 tested GPT Image 2.5 with a bare prompt — 'make me a realistic iphone photo of a woman in a cafe' — and got a result realistic enough to pass as a real photo. The author says they don't understand how people get bad results from the model, given how few words it takes. A casual user test of the image model's photorealism.
More from Multimodal
- VoiceStudio open-sources a local ElevenLabs-style stack: voice clone, dubbing, 646 languages — thisguyknowsai · 2026-09-09
- OpenAI image-2.5 hands-on: sub-$0.01, 23s generations clearly beat Muse — tobowers · 2026-09-09
- GPT image model generates a 10-second GIF, sparking video-model takeover talk — rounak · 2026-09-09
- TapNow accuses influencer of stripping watermark to pass off signed artist's SD 2.5 AI film as rival product demo — TomLikesRobots · 2026-09-09
- Marigold V2 retools diffusion transformers for sharper monocular depth estimation — huawei-bayerlab · 2026-09-09
- One-Take Rainy Toll Booth: A Full AI Video Prompt for Korean Supernatural Mystery — umesh_ai · 2026-09-09