ChatGPT Images 2.0 is better at rendering non-Latin text - Engadget
ChatGPT Images 2.0 is better at rendering non-Latin text OpenAI describes it as a "step change" for image generation models. By Igor Bonifacic April 21, 2026 3:00 pm EST…
Search
All sources
huggingface.co
23
theverge.com
15
engadget.com
12
androidauthority.com
6
wired.com
6
techcrunch.com
5
cnet.com
5
androidpolice.com
4
9to5mac.com
4
arstechnica.com
3
techspot.com
2
neowin.net
2
People also ask
What’s Different?
Since the new model can tap into ChatGPT’s “reasoning” capabilities, Images 2.0 can search the internet for recent information and generate more than one image at a time. In essence, the bot can use additional steps to output more thorough generations from a single prompt. Images 2.0 also has a more recent knowledge cutoff date: December 2025. This also means that outputs from the new model are more granular. For example, I generated an infographic with San Francisco’s weather forecast for the next day, as well as activities worth doing. The image ChatGPT generated included accurate weather de
OpenAI Beefs Up ChatGPT’s Image Generation ModelVideos
More videos
Inside image generation’s Renaissance moment — the OpenAI Podcast Ep. 19
Why AMD Laptops Are Perfect for Local AI Image Generation (with Render FX)
ownCloud Is Everything Nextcloud Should Have Been
Introducing Gemini Omni: Create Anything from Anything
Portainer to Dockhand: One Container Replaced Several
What is Gemini Omni?
A New VOID RUN Mission | Build Local AI with Ryzen™ AI Halo
The Best 'Not A Chatbot' Software from NVIDIA GTC 2026 - Camb.ai and SkillReal
Energy-Based Transformers are Scalable Learners and Thinkers (Paper Review)
What the Freakiness of 2025 in AI Tells Us About 2026
Top stories
Gemini can now draw on your Google data to personalize the images it generates - Engadget
Gemini can now draw on your Google data to personalize the images it generates Your Google data can act as an extra reference point for Nano Banana 2. By Ian Carlos Campbell…
Google's Circle to Search feature can tell you if an image was AI-generated - Engadget
Google's Circle to Search feature can tell you if an image was AI-generated The company is expanding SynthID to Chrome and Search. By Karissa Bell May 19, 2026 1:45…
AI Image & Video | Tom's Guide
…AI Image & Video Google has upgraded its AI image generation lineup with Nano Banana 2, a brand-new model designed to deliver studio-quality visuals at blistering speeds. AI Image & Video AI…
Discussions and forums
Pixel-Aligned 3D Generation from Images
Pixel-Aligned 3D Generation from Images
1
Show HN: PhoneDiffusion – Local AI image generation for iOS
PhoneDiffusion is officially liveNow you can generate AI images locally on iPhone with Stable Diffusion. Fast, private, account-free and specifically built for iOS.What you can do: ∙ Generate images in under 5 seconds on…
11
Pixal3D: Pixel-Aligned 3D Generation from Images
Pixal3D: Pixel-Aligned 3D Generation from Images
2
Pixal3D: Pixel-Aligned 3D Generation from Images
Pixal3D: Pixel-Aligned 3D Generation from Images
2
Show HN: Run TRELLIS.2 Image-to-3D generation natively on Apple Silicon
I ported Microsoft's TRELLIS.2 (4B parameter image-to-3D model) to run on Apple Silicon via PyTorch MPS. The original requires CUDA with flash_attn, nvdiffrast, and custom sparse convolution kernels: none of which work …
202
40
Gemini can now pull from Google Photos to generate personalized images
…Google Gemini Personal Intelligence image generation demo] Google notes that if you do opt in to Personal Intelligence, the company will not “directly train” its AI models on your private Google Photos…
OpenAI says it’s getting serious about AI detection and labeling
…When an image is uploaded, the portal will check C2PA and SynthID provenance signals to flag if it was generated with ChatGPT, the OpenAI API, or Codex. This is limited to images…
The 70-Person AI Image Startup Taking on Silicon Valley's Giants
In December, Black Forest Labs raised funds at a $3.25 billion valuation, after signing deals to power AI image-generation features in Adobe and the graphic design platform Canva. It has…
Has Google’s AI watermarking system been reverse-engineered?
A software developer claims to have reverse-engineered Google DeepMind’s SynthID system, showing how AI watermarks can be stripped from generated images or manually inserted into other works. A claim that…
Sony and TSMC are teaming up on image sensors.
…Sony, TSMC plan new Japan joint venture for next-generation image sensors | https://www.reuters.com/world/asia-pacific/sony-tsmc-plan-new-japan-joint-venture-next-generation-image-sensors-2026-05…
xAI is being sued by teens who say Grok created CSAM using their photos - Engadget
…xAI CEO Elon Musk, who previously promoted Grok's "spicy" abilities, has claimed that he was "not aware of any naked underage images generated by Grok." xAI announced in January it would…