OpenAI just rolled out GPT Image 1.5, now the default image engine inside ChatGPT, and it’s a meaningful step forward, not just a version bump.
The headline change is the new Images workspace: presets and styles (VHS, Watercolor, Cyberpunk, K-Pop), a discovery feed with trending prompts, and an image library. This alone makes ChatGPT feel closer to a real creative tool, not just AI Slop Image Gen.
I put GPT Image 1.5 head-to-head with Google’s Gemini image model (Nano Banana Pro) across 10 practical tests, including:
Precise image editing (preserving lighting, faces, and composition)
Generation speed
Text rendering (posters, newspapers, UI)
Complex prompt following
Character consistency across multiple images
Logo design
Real-world context (Tesla, SpaceX)
Famous people generation + editing
Website UI mockups
Style transfers (Minecraft, Ghibli)
What stood out:
GPT Image 1.5 is much better at editing existing images without breaking realism.
Text rendering is finally reliable enough for posters and layouts.
Instruction following is noticeably stronger on long, multi-constraint prompts.
The new one-time likeness upload is great for consistent characters.
Speed is improved, still not the fastest overall, but no longer painfully slow.
Gemini remains competitive, especially on fast generations and some stylized output, but OpenAI is clearly optimizing for creative workflows, not just raw output.
I walk through all 9 tests, side-by-side results, and where each model wins in the demo video below.
👉 Watch the full comparison and decide which image model actually earns a spot in your stack.
Thanks for watching!
🤖 If you want to stay ahead of the curve on all things AI, make sure to join our community across all platforms:
📩 Get the Weekly Newsletter: https://thisweekinai.ai/
📺 Subscribe on YouTube: www.youtube.com/@ThisWeekinAIPodcast
📸 Instagram: www.instagram.com/thisweekinaipodcast
📱 TikTok: www.tiktok.com/@thisweekinaipodcast
