OpenAI Unveils ChatGPT Images 2.0 to Challenge Google’s Gemini Nano Banana 2

In early 2025, OpenAI released a new image‑generation model that quickly captured public attention and attracted millions of fresh users to its flagship chatbot. By April of the same year, the company extended this technology to developers via the gpt-image-1 API, and later introduced an upgraded version, gpt-image-1.5, in December 2025.

Google’s Gemini Nano Banana Series Continues to Expand

Since September of last year, Google has been deploying its Gemini Nano Banana line of image‑generation models. This year, the firm announced Nano Banana 2—also known as Gemini 3.1 Flash Image—which offers “Pro‑level” visual quality and notable enhancements over earlier releases.

OpenAI’s New Offer: ChatGPT Images 2.0

To compete with Gemini Nano Banana 2, OpenAI introduced ChatGPT Images 2.0 during a livestream featuring CEO Sam Altman and other executives. The new model excels at producing images that contain text, enabling accurate rendering of elements such as macOS desktop windows or chat interfaces.

Key capabilities include:

  • Closer adherence to user instructions and preservation of requested details
  • Precise depiction of fine‑grained components—small text, icons, UI widgets, dense compositions, and subtle stylistic cues
  • Resolution up to 2 K across aspect ratios ranging from 3:1 (wide) to 1:3 (tall)

Dual Model Variants

OpenAI offers two versions of Images 2.0:

  • ChatGPT Images 2.0 instant
  • ChatGPT Images 2.0 thinking

The “thinking” or Pro model can query the web for real‑time information related to a prompt, generate multiple distinct images from a single request, and verify its own outputs.

Improved Multilingual Support

Images 2.0 demonstrates stronger multilingual understanding, particularly with non‑Latin scripts such as Japanese, Korean, Chinese, Hindi, and Bengali.

Developer Access via the API

The gpt-image-2 model is available to developers under the following pricing structure:

  • $8.00 per input prompt
  • $2.00 for cached input reuse
  • $30.00 per output image

User Availability

The instant variant is accessible to all ChatGPT and Codex users, while the thinking variant is limited to ChatGPT Plus, Pro, and Business subscribers.

Whether you need a small assistant for one team or a full agentic AI workflow for the whole company, we size the setup to what you need and what your team can manage. Get in touch and we’ll map it out with you.

Chat with AI

Hello! I'm MTLabs AI, How can I help you today?