ChatGPT Images 2.0 is here: what launched, what the API exposes, and what to test now
On April 21, 2026, OpenAI published the ChatGPT Images 2.0 system card and the `gpt-image-2` API model page. This site now focuses on what actually shipped: stronger world knowledge, denser text rendering, better instruction following, image editing, flexible sizes, and the practical differences between ChatGPT Images 2.0 and the `gpt-image-2` API.
How to read ChatGPT Images 2.0 correctly
The clean way to read this launch is to separate three layers: ChatGPT Images 2.0 as a product experience, `gpt-image-2` as the API model, and the image workflows you want to benchmark on Vofy.
Official Release
OpenAI published the ChatGPT Images 2.0 system card on April 21, 2026 and documented the `gpt-image-2` API model the same day.
API Model
The official model page describes `gpt-image-2` as OpenAI's state-of-the-art image generation model for fast, high-quality generation and editing with flexible sizes and high-fidelity image inputs.
Product vs API
ChatGPT Images 2.0 adds a product-layer experience including Thinking mode and transparent-background workflows in ChatGPT, while the API docs say `gpt-image-2` does not currently support `background: "transparent"`.
Start from the April 21 launch sources
Use the OpenAI system card, model page, image guide, pricing page, and Help Center article as the baseline instead of outdated rumor-track content.
Separate ChatGPT from the API
Do not flatten the launch into one generic claim. ChatGPT Images 2.0 and `gpt-image-2` overlap, but availability, transparent backgrounds, and Thinking mode differ by surface.
Benchmark the jobs that matter to you
Use Vofy to compare product edits, dense-text layouts, poster-like graphics, and multi-style prompt sets against the new official baseline.
Track costs, sizes, and QA together
The `gpt-image-2` docs now include flexible sizing, pricing guidance, rate limits, moderation behavior, and snapshot naming. Treat those as part of the evaluation, not as afterthoughts.
What teams will care about most
The official launch matters because it improves the parts that affect real creative work: editing fidelity, dense text and multilingual layouts, and more structured control over visual outputs.
Better image editing
The official API model is positioned around fast, high-quality image generation and editing with high-fidelity image inputs, which makes product edits and controlled revisions a central use case.
Dense text and multilingual layout
OpenAI's launch materials emphasize dense text, readable typography, multilingual rendering, and more layout-heavy editorial outputs than the previous generation.
More structured control
The launch matters because users want stronger instruction following, richer world knowledge, and more reliable handling of complex image briefs instead of loosely interpreted prompts.
Common questions about ChatGPT Images 2.0
These are the questions most likely to matter before someone wires ChatGPT Images 2.0 into a real workflow or benchmark plan.
Is GPT Image 2 official?▼
Not as a formal product name. OpenAI officially launched ChatGPT Images 2.0 on April 21, 2026, and the API model is documented as `gpt-image-2`. “GPT Image 2” is still better understood as a shorthand or search term.
Is GPT 2 Image the same thing?▼
Usually it is just a search variant for ChatGPT Images 2.0, but it is not the official product name. The official public names are ChatGPT Images 2.0 and `gpt-image-2`.
Does the API support transparent backgrounds?▼
Not currently. The image generation guide says `gpt-image-2` does not currently support `background: "transparent"`, even though the ChatGPT product help article says ChatGPT Images can make the background transparent.
Why does this page point to Vofy?▼
Because Vofy gives you a practical place to benchmark the kinds of tasks this launch targets: editing, text-heavy layouts, poster systems, visual variants, and model comparison work.