OpenAI is setting new standards in the technology world with GPT-4, an AI that can now also interpret images
OpenAI, the company behind some of the most advanced AI models, has unveiled its latest development: GPT-4. This model represents a significant advance in the world of artificial intelligence, as it can not only understand and generate texts but can now also analyze images.
Innovative image recognition
Although GPT-4’s ability to process images is not yet available to all users, impressive functions have already been demonstrated. For example, users can upload photos of objects and receive suggestions as to which recipes are possible with the ingredients shown or which irregularities an image has.
Wide range of applications
The possible applications of GPT-4 are enormous and range from supporting people with visual impairments to creating websites from sketches to writing creative texts and programming aids. For example, it is possible that with GPT-4, users can generate a working website from a simple sketch, even if they do not have in-depth programming experience.
More capacity, more creativity with GPT-4
A key upgrade is the expanded capacity of GPT-4. The model can now work with up to 25,000 words, a significant increase over the 3,000 words of its predecessor, GPT-3. This improvement allows more complex requests to be processed and longer content, such as almost complete books, to be created.
Academic excellence
In academic testing, GPT-4 shows amazing results that are not only an improvement over GPT-3.5 but also rival the performance of subject matter experts.
Usage and availability
Those interested can access GPT-4 via the ChatGPT Plus subscription, which is available for a monthly fee of around 20 euros. Here, users can choose between different versions of the model, with GPT-4 being slightly slower than its predecessor but significantly more intelligent.
Bing Chat as an alternative
Microsoft offers a free alternative to testing GPT-4 with the new Bing Chat, which is already accessible to many users or can be reached via a waiting list. Here, GPT-4 is optimized for search queries, which enables shorter and to-the-point answers.
First impressions and tests of GPT4
In initial tests, GPT-4 shows considerable creativity and problem-solving ability. The model shows its strengths even with complex puzzles or creative challenges, such as writing texts where each word begins with a specific letter.
OpenAI has taken a huge step forward with GPT-4. While not all features have been activated yet, such as image analysis and the processing of longer text contexts, the potential of GPT-4 is already clearly visible. This update is a significant milestone in the evolution of artificial intelligence and will certainly lead to a variety of innovative applications.