The Best New Image Model In The World
GPT Image 2 has been declared the top image generation model, with OpenAI announcing it entirely through images on their website. The video showcases several impressive demos including scannable barcodes, comic book aesthetics, detailed cell biology diagrams, and dense readable newspaper text.
Summary
The video covers the launch of GPT Image 2, which the presenter declares the best new image model in the world. Notably, OpenAI announced the model on their website using images exclusively — the entire announcement was written on top of images rather than standard text.
Several impressive user-generated demos are highlighted. Riley Brown demonstrated that GPT Image 2 can generate a realistic book cover for 'Good to Great' complete with a functional barcode — when scanned with a phone, the barcode actually redirected to the real book listing, showcasing the model's unprecedented accuracy with embedded data.
Justine Moore shared a comic book-style image that the presenter praised for its strong aesthetic quality and vibrant color palette. Bo Wang generated a detailed diagram of a biological cell, including labeled structures like the nucleus and plasma membrane, which the presenter found visually convincing even while admitting limited biology knowledge.
Finally, Mark generated an image of a newspaper filled with dense, coherent, and readable text — an area historically difficult for image generation models. The presenter noted the text appeared accurate and legible, highlighting this as a significant leap in capability.
Key Insights
- OpenAI announced GPT Image 2 entirely through images on their website, with the full announcement written on top of images rather than standard webpage text.
- Riley Brown demonstrated that GPT Image 2 generated a barcode on a book cover that was functionally scannable and redirected to the actual 'Good to Great' book listing.
- Justine Moore's comic book image was highlighted by the presenter specifically for its strong color palette and aesthetic quality, suggesting the model has improved stylistic coherence.
- Bo Wang's generated cell diagram included labeled biological structures like the nucleus and plasma membrane, impressing the presenter despite his self-admitted limited biology knowledge.
- Mark's AI-generated newspaper image contained dense, coherent, and fully readable text, which the presenter suggested may even be factually accurate — a historically weak area for image generation models.
Topics
Transcript
[0:00] GPT image 2 is now the top dog. Interestingly, their website announced it fully in images. The whole thing is written up on top of images. My buddy Riley Brown here posted this video on X. He asked it to generate an image of the book. Good to great. The barcode on the book should scan to the actual book. Riley then used a barcode scanner on his phone here to scan the barcode. And look at that. It actually took him to the book. Good to great. Like that was super impressive. Justine Moore here showed this comic book image. I really like the [0:32] aesthetic. Like the colors look really, really good. Bo Wang here showed an…
Full transcript available for MurmurCast members
Sign Up to AccessMore from Matt Wolfe
AI News: Dots, GPT-6.1 Sol, Sonnet 5.5, Gemini 4, and everything you need to know
A comprehensive review of major AI announcements from the week, including OpenAI's new Dots agent, GPT-6.1 Soul model, Anthropic's Sonnet 5.5, and Google's Gemini 4 Argon, with analysis of pricing, capabilities, and competitive positioning across different models.
The Hands Down Best Coding Model Right Now
The speaker reviews Opus 5.5, claiming it's currently the best state-of-the-art coding model at a reasonable price point. They demonstrate its capabilities by running a 20-hour test where the AI built a nearly complete recreation of the Mega Bonk game, including accurate character tiers, sound design, and UI elements.
This AI Model Does’t Use Words?!
Typesafe AI released Jev, a new AI model that outputs structured decisions (choices, scores, or booleans) instead of generated text like traditional LLMs. This approach makes Jev dramatically cheaper and faster, with input costs at 4 cents per million tokens and free output tokens.
AI News: Opus 5.5, GPT-6 Sol, Jev, Muse and More!
This week in AI saw major announcements from Meta Connect (Muse agent, new AR/VR glasses), three new language models (Claude Opus 5.5, GPT-6 Soul, Grok 4.7), significant buzz around Typesafe AI's Jev decision model, and new features from YouTube, Microsoft, Google, and Spotify. Claude Opus 5.5 emerged as the new state-of-the-art model while Jev introduced a fundamentally different approach to AI outputs focused on decision-making rather than text generation.
5 Ways To Use ChatGPT's New Features
The transcript demonstrates five practical applications of ChatGPT's new features, including computer use for software automation, advanced image editing, cloud browser integration for account access, voice-powered live agents, and persistent session agents for long-term user interaction tracking.