OpenAI Drops ChatGPT Images 2.0 that Can Generate Text in Hindi & Bengali
The new ChatGPT Images 2.0 tool introduces enhanced flexibility, allowing users to generate images up to 2,000 pixels wide in a wider range of aspect ratios.
OpenAI has launched ChatGPT Images 2.0, a major upgrade to the image-generation capabilities built into its chatbot, alongside a new enterprise-focused training initiative called Codex Labs.
The updates signal the company’s continued push to expand its footprint across both creative AI and developer ecosystems.
The new ChatGPT Images 2.0 tool introduces enhanced flexibility, allowing users to generate images up to 2,000 pixels wide in a wider range of aspect ratios. This includes formats suited for infographics and design-heavy content, addressing a key limitation in earlier versions.
The company also highlighted significant improvements in image quality, particularly in text rendering. The upgraded model can more accurately generate multilingual text, including Japanese, Korean, Chinese, Hindi and Bengali.
Beyond language capabilities, the system has improved its handling of small visual elements such as icons, interface components and fine details. OpenAI noted that the tool now produces images with subtle imperfections that add realism, reducing the overly polished look often associated with AI-generated visuals.
ChatGPT Images 2.0 also leverages a built-in knowledge dataset to interpret prompts more intelligently, filling in missing details when needed. Paid users can further enhance outputs by enabling advanced reasoning modes that draw on public web data, improving both accuracy and design structure. The tool can generate up to 10 images per prompt, streamlining workflows for creators and designers.
Alongside this release, OpenAI introduced Codex Labs, a new service aimed at helping enterprises adopt its Codex programming assistant through training, workshops and integration support.
“Our professionals are using Codex to move from static requirements to working solutions in hours, not weeks. It’s enabling rapid prototyping, real-time workflow redesign, and faster iteration across the development lifecycle. That speed translates directly into faster builds and better outcomes for our clients,” Lan Guan, Accenture Chief AI Officer, said.
