The new features of GPT-4V(ision) have been used in various ways, including turning mockups into live websites and code, identifying objects, Optical Character Recognition (OCR), interior design, identifying movies, turning food pictures into recipes, helping with studies, decoding infographics, charts, and troubleshooting and teaching software. OpenAI has not shared the details but said that image and voice features will also be made available for the free users of ChatGPT in the future.
Key takeaways:
- OpenAI has introduced new image and voice features for GPT-4, making it their most advanced multimodal model, GPT-4V(ision). It can now accept multiple modalities of input, including text, images, and voice.
- GPT-4V(ision) has been gradually rolling out to Plus and Enterprise subscribers of ChatGPT since its launch announcement. It is available on all platforms, including web, iOS, and Android.
- Some of the most popular use cases for GPT-4V(ision) include turning mockups into live websites and code, identifying objects, Optical Character Recognition (OCR), interior design, identifying movies, turning food pictures into recipes, helping with studies, decoding infographics, charts, and troubleshooting and teaching software.
- OpenAI has not shared the details but said that image and voice features will also be made available for the free users of ChatGPT in the future. GPT-4V(ision) will also be made available for developers through APIs at a later date.