The Open-Source AI Challenging ChatGPT

DeepSeek-R1 is a new AI reasoning model from the Chinese company Deepseek.

Released Jan 20 offers a cost-effective alternative to chatgpt.

🚀 DeepSeek-R1 is here!

⚡ Performance compared to Openai-O1
📖 Fully open model and technical report
🏆 MIT licensed: distill and commercialize freely!

🌐 The website and API are now live! Try deep into https://t.co/v1tfy7lhny today!

🐋 1/n pic.twitter.com/7blpwapu6y

– DeepSeek (@deepseek_ai) January 20, 2025

Here’s why the DeepSeek-R1 is currently trending around the web.

Key features

Thinking like a human

DeepSeek-R1 has advanced reasoning skills that help solve complex problems in math, logic and coding.

People praise its ability to mimic human thinking. Breaks down problems into smaller steps using a “chain of thought” (COT) method.

As it processes its responses, DeepSeek-R1 can adjust responses in real-time and experience “AHA” moments while solving tricky problems.

Here’s a screenshot from Deepseek’s research paper (PDF link) Showing where this moment happened:

Here’s another screenshot more representative of what you’ll likely see when using the web interface. This is DeepSeek’s thought process when presented with an SEO question:

His train of thought continued for a number of paragraphs before he finally formed an answer.

Open source

DeepSeek-R1 is an open source model published under MIT licensewhich means that anyone can use and modify their code.

This openness makes DeepSeek-R1 attractive to companies, startups and developers looking for affordable AI solutions.

Lower development costs

While companies like OpenAi spent hundreds of millions to develop their models, DeepSeek-R1 allegedly Built on a budget of just $6 million.

DeepSeek achieved this by using data more efficiently and applying reinforcement learning strategies.

This economy was achieved by optimizing the use of data and applying reinforcement learning strategies in a novel way that departed from the conventional supervised fine-tuning processes commonly used to train large language models.

This reduced the need for large amounts of computing power, making it more affordable for end users.

Affordable prices

DeepSeek-R1’s competitive pricing is another factor contributing to its growing popularity.

It is completely free to use through chat.deepseek.com. And if your machine has the necessary specifications, you can also run the model locally on your computer at no cost.

For those without such resources, DeepSeek offers a cloud-based API service at prices well below industry standards.

In addition, DeepSeek offers a cloud-based API service. Accessing the model through this API has costs, but prices it is significantly lower than many competitors.

Is that good?

Although DeepSeek-R1 is praised for being affordable and open source, opinions on its performance differ.

Many reference values Show it performs with OpenAi’s O1 model in areas such as logical reasoning and problem solving.

While DeepSeeK-R1 may have unprecedented limitations, it is a useful option for tasks that require systematic, step-by-step reasoning.

Its open source nature allows for rapid iteration, making it a dynamic and evolving tool.

What people are saying

The release of DeepSeek-R1 sparked widespread debate about its potential to democratize access to AI.

Launching the model also carries geopolitical significance.

Analysts see DeepSeek-R1 as a demonstration of China’s progress in AI, especially in light of US controls on technology exports.

By achieving competitive results with a fraction of the resources, DeepSeek highlights the growing global competition in AI.

Community reactions

Here’s a preview of the debates you may have missed over the weekend:

DeepSeek R1 is one of the most amazing and advanced breakthroughs I’ve ever seen – and as open source, a profound gift to the world. 🤖🫡

– Marc Andreessen 🇺🇸 (@PMarca) January 24, 2025

Goodbye chatgpt

It’s only been 5 days since the Deepseek R1 dropped and the world has already blown its potential.

13 examples that will blow your mind (don’t miss the 5th): pic.twitter.com/u5yelfgxam

– Poonam Soni (@codeBypoonam) January 25, 2025

American Tech Bros are suing a company from China, they released a new AI model called Deepseek and it’s much better and efficient and it only cost them 6 million dollars to make compared to Americans who spend billions.. and they made it open source, so so that everyone can look at the code and make their own🙃 https://t.co/q9mzu4w90w

—TOri wan kenobi (@majestyria) January 25, 2025

Deepseek stole Ai Thunder:

– with zero hype from the CEO,
– zero “OMG guys change it every” influencers ”
– There are no easy demonstrations
– No inflated promises
– No hints on “AGI achieved internally”

They did it

ships the actual product. pic.twitter.com/qdzmqycq7t

– Michael Kove (@michael_kove) January 20, 2025

So wait wait, the founder of Deepseek is basically Jim Simons from China and he was doing this LLM thing just as a side project, and for $6 million he managed to detroy every AI company in the world.

That’s how we are cooked, hahaha pic.twitter.com/qoig1bxj20

- Q-cap  (@qcapital2020) January 24, 2025

I asked #R1 to visually explain the Pythagorean theorem.

This was done in one frame without mistakes in less than 30 seconds.

Wrap it up, done: #Deepseek #R1 pic.twitter.com/nyjjslk7if

– Christian H. Cooper (@Christiancooper) January 20, 2025

Looking ahead

DeepSeek-R1 represents a milestone in the AI ​​race, offering a highly efficient, cost-effective alternative to established tools.

While it may still not outperform its competitors in every aspect, its affordability and affordability position it as a transformative tool for many applications.

Broader impact on the market

The release of DeepSeek-R1 is influence Global markets, especially in AI and technology. After its launch, tech stocks fell sharply as investors questioned the need for big hardware investments.

Nvidia, for example, lost over $300 billion in market value, the largest one-day loss for any company.

This is a developing story…