ChatGPT has not really had a competitor so far, but it may have found a worthy challenger in DeepSeek.

Developing AI tools costs a lot of money, which is why emerging startups usually seek support from big tech companies. DeepSeek emerged almost from nothing.

DeepSeek’s R1 model officially advertises itself as a competitor to o1. Many claim that the Chinese chatbot can be a worthy challenger to OpenAI’s most popular model.

An interesting fact is that training the model cost much less ($5.6 million instead of $100 million) and using the API is also significantly cheaper.

According to the MIT technology review:

“the sanctions appear to be driving startups like DeepSeek to innovate in ways that prioritize efficiency, resource-pooling, and collaboration.”

This is a very interesting statement, because if efficiency increases and a system can be trained faster and cheaper, then it can be assumed that its energy consumption will also be lower.

Even the stock market felt the advance of DeepSeek: Nvidia’s share price fell significantly.

Registration

Registration is very simple: you can even log in with a Google account and the service is currently completely free.
Although unfortunately, at the time of writing, logging in is quite difficult. According to the company, the system is currently under attack by a large-scale hacker.
Update: It has since been revealed that DeepSeek mobile applications have already been banned in Italy due to data protection issues, and the France’s privacy watchdog has launched an investigation into it, so there is a lot of interest in the new chatbot in all respects.

Let’s see what it can do

Test #1 – Open Question

After logging in, I started with a simple open question:

I then asked what is the situation now, in 2025?
It also gave a detailed answer to this and briefly outlined possible future developments in the field of AI image generators (e.g. DALL-E 4, further Stable Diffusion versions, and so on)

Test #2 – Searching Website Content

I then followed up with a question that required a bit more analysis:
“What is the latest post in the AI ​​Timeline on aiartblog.com?”

To which it replied that it had no access to the internet:

When using DeepSeek’s Search feature, I received this message:
“Sorry, DeepSeek search service is busy. Please disable search or try again later.”

ChatGPT gave me the correct answer:

Test #3 – Pythagorean theorem

The next test was a medium difficulty math problem: explaining the Pythagorean theorem. I was curious to see how it would display the formulas and if it would draw a diagram.

Here’s how DeepSeek AI solved it (excerpt from the answer):

ChatGPT asks if we want visual proof, and then asks if it should generate an image for it.

The explanation is detailed, but the image is unfortunately not very useful:

Test #4 – Creating a 3D model from an AI image

In the following, I was curious how it would answer this question:
“How can I create a 3D model from an AI generated image?”

It also gave a detailed answer here, but only indicated the references in the text after being asked:

Test #5 – Image Generation

Finally, I asked it to draw a picture using a previous ChatGPT prompt I wrote for an article about DALL-E 3 going free.

The prompt: 

“Imagine an oil painting of Yggdrasil, the immense and mystical ash tree that connects the Nine Worlds in Norse mythology. The tree stands tall and grand, its sprawling branches reaching up to the heavens, filled with lush green leaves that shimmer with an otherworldly glow. The trunk is thick and gnarled, with intricate patterns etched into the bark, symbolizing the ancient runes of wisdom.

At the base of Yggdrasil, the roots dig deep into the earth, entwining around sacred wells like the Well of Urd. The roots are surrounded by mist, hinting at the connection to the underworld. The scene captures the tree as the cosmic axis, with the sky above painted in deep blues and purples, speckled with stars, and the ground below showing hints of the realms connected by the roots.

A serpent can be seen coiling around one of the roots, representing Nidhogg gnawing at the base of the tree, while an eagle perches high in the branches, its sharp eyes gazing out over the worlds. Other creatures from Norse myth, like the stags that graze on Yggdrasil’s leaves, are subtly included in the composition, adding layers of depth and storytelling.

The painting exudes a sense of ancient power, mystery, and the interconnectedness of all things in the Norse cosmos, with vibrant colors and rich textures that convey the majesty and sacredness of Yggdrasil.”

And the result:

So DeepSeek can’t generate an image directly in the chat window. But they released the Janus-Pro-7B image generator which is downloadable from GitHub or you can try it out on Huggingface.

It is maybe not the challenger of Dall-E, but it will be better shortly I think.

I generated this test images on Huggingface. Click on the thumb block to see the full sized images on Facebook.

It’s clear that the image quality is still lagging behind its big rivals. The model still has a lot to learn here.

My summary

DeepSeek is a promising chatbot that could be a worthy challenger to ChatGPT. It is said to heavily censor results on certain topics (such as Chinese historical and political events). Fortunately, this does not affect us in terms of AI art.
Based on my tests, I consider o1 to be an even better model, but it is a fact that DeepSeek is also noteworthy.
Unfortunately, DeepSeek’s servers are very busy, so its use can be difficult during peak hours.
The news about data protection that has been appearing in recent days is thought-provoking, and it is worth reading them.
Considering that training R1 took much less energy and this chatbot is quite easy to use, it is worth trying out what it is capable of.

Advice: I would not give it confidential data considering the ongoing data protection investigations and previous news about how Chatbots use the data provided to them.

Cover image based on:
Image by Gerd Altmann from Pixabay

The post I tested DeepSeek vs ChatGPT with 5 prompts appeared first on AI Art Blog.

In this article, we’ll take a look at what OpenAI’s latest GPT-4o model can do. If you’re curious about its capabilities, be sure to read on.

On Monday, OpenAI announced its newest, fantastic model named GPT-4o. They practically wrote an article and there are plenty of videos on their YouTube channel in which they demonstrate its capabilities.

We will go through them in broad outline, so that by the end of the article you will have a comprehensive picture of what makes this new large language model (LLM) so groundbreaking.

It is noteworthy that GPT-4o will also be available for free.

News about GPT4-o

Its most notable ability is that you can practically have a conversation with it. The reaction time for answering has been reduced to a minimum with this AI model, and you can also interrupt his speech.

The previously available Voice service of ChatGPT needed 4-5 seconds for the answer, now that has completely changed.

Watch this example video to see what’s new.

The GPT4-o model

In the description, you can read that this model is actually an omni model. This means, that it can not only interpret your voice (Voice mode) and text, but can practically handle both.

The interpretation of images and videos is made possible by the Vision mode.

The Memory mode remembers our previous conversations, which ChatGPT can use later. With the Analyze mode, complex data can be analyzed quickly and efficiently. Examples: Excel spreadsheets or even a chart that you show as an image to chatGPT.

Practical examples

If you look at the introduction page of GPT-4o, you can see many examples of people using their phone’s camera with the ChatGPT app and asking ChatGPT what it sees in the camera, and it then answers them “verbally”.

So the same thing can now be done in real time with your phone’s camera. The big difference between the current GPT-4o and the previous Voice service is that a completely new model was actually trained here with voice, text, videos and images.

It is also worth watching the video where they show that one person speaks Italian, the other person speaks English. The phone interprets it in real time. This means that as soon as the Italian text is spoken, ChatGPT immediately translates it into English and vice versa.

You’ve may seen conversations, where politicians spoke in their own mother tongue and still understood each other. It was thanks to the dubbing service in the background.

In the future, this will probably be completely replaceable with such an AI solution.

Performance of GPT-4o

There is a comparison chart on the hello-gpt-4o page in the Model evaluations section, with which we compare the GPT-4o model through various benchmarks, for example with GPT4 turbo, or even with the paid Claude 3 Opus and Gemini Pro 1.5.

The speed difference according to the graph will become visible in practice. All models are already very fast and are therefore capable of responding in real time.

You can feel the difference between a free ChatGPT, i.e. GPT 3.5 and GPT 4, but the difference between the above big models is not really noticeable anymore.

Security and limitations

The good news is that they also put a lot of emphasis on security. It’s highly likely that you won’t be able to say obscene words with it and you won’t be able to ask ChatGPT for other silly things either.

This model is subjected to serious, in-depth testing, which means that a team tries to hack these systems and extract answers from them. The name of the testing method is the so-called red teaming.

The future

These services are constantly being improved. Maybe by the time you read this article, the entire interface will look completely different.

Thanks to the Memory mode, ChatGPT will also remember previous conversations. You can turn this off if you don’t want it to store information about you.

The Analyze mode is good for uploading documents and requesting all kinds of analytics from it. So there will be improvements here in the free GPT as well.

Summary

This new model contains huge innovations. It almost feels like talking to a live person. It can also be used to facilitate scientific tasks, such as the analysis of complex data, graphs, and images.

If this is also combined with image and video generation, or perhaps with the generation of sound and music, the limit to what we can create with it is the imagination.

Of course, he won’t cook lunch for you, but he can help you in creative and scientific tasks.

Thank you for reading this article!

Image by Franz Bachinger, Pixabay

The post OpenAIs GPT-4o model is free to use! appeared first on AI Art Blog.