Chinese AI startup DeepSeek has announced it will launch five new open-source repositories starting next week, marking what the company is calling its “Open-Source Week.” The announcement, made on Friday via the company’s X account, underscores DeepSeek’s commitment to transparency and collaboration as it pushes the boundaries of Artificial General Intelligence (AGI).
“We’re a tiny team at DeepSeek exploring AGI,” the company stated in its post. “Starting next week, we’ll be open-sourcing five repos, sharing our small but sincere progress with full transparency.” This bold move reinforces DeepSeek’s reputation as a leader in cost-effective, high-performance AI.
It challenges industry giants like OpenAI with models such as DeepSeek-R1 and DeepSeek-V3, which have received widespread acclaim for their performance at a significantly lower cost than their Western counterparts.
Currently, DeepSeek offers a strong portfolio of 14 open-source models and repositories available on Hugging Face, including its latest models like DeepSeek-R1, which rivals OpenAI’s o1 model in reasoning tasks, and DeepSeek-V3, trained for under $6 million a stark contrast to the hundreds of millions spent by competitors.
Nvidia’s market share declined by around $589 billion in one day, raising questions about the amount of capital and resources required to build powerful AI models.
2048 NVIDIA H800 GPUs used by DeepSeek to train DeepSeek-V3 to get performance, beating all open sources and closed sources. Andrej Karpathy stated the DeepSeek-V3’s level of capability is “supposed to require clusters of closer to 16,000 GPUs”.
For Example, xAI’s new Grok 3 (world’s smartest AI model) has been trained on over 100k NVIDIA GPUs and it is slightly better than DeepSeek-R1, I have compared the ai model with competitors.
These repositories can contain key company assets, the main reason for making free DeepSeek models, any new rising start-ups can modify and leverage according to usage, and if API is needed that will be affordable. The company states “As part of the open-source community, we believe that every line shared becomes collective momentum that accelerates the journey“.
Posts on X reflect excitement and anticipation, with users hailing DeepSeek’s move as a “democratization of AI” and a potential game-changer for the open-source model.
On the other hand. Sam Alman against open source models and to counter this statement said “I think we should probably open source somewhat more” in a recent interview with Sky News.
This extended to the Reddit ask-me-anything discussion he said “I personally think we have been on the wrong side of history here and need to figure out a different open source strategy; not everyone at Openai shares this view, and it’s also not our current highest priority“.
I’m David Paul, a news writer at DeepSeek Insider, where I cover the latest developments in artificial intelligence, large language models (LLMs), and emerging AI tools in the market. With a keen interest in these cutting-edge technologies, I deliver insightful and timely updates to keep our audience informed and engaged.