Lately there is a lot of talk about the big AI models, but there is a parallel trend that is gaining a great deal of momentum: open-source or ‘open weights’ models. Companies such as Google, Microsoft, Alibaba and Nvidia are launching smaller but incredibly efficient models, like Gemma 4 or Qwen 3.5, that are revolutionizing the business landscape.
The great advantage of these models is that companies can run them on their own servers without having to spend fortunes or compromise the privacy of their data by sending it to third-party clouds. According to industry experts, you no longer need the biggest and most expensive model on the market to solve everyday problems; sometimes a more modest but well-optimized model is exactly what you need. Google’s Gemma 4, with 31 billion parameters, is already the fourth highest-rated open model in the world.
This democratization of AI is fantastic news. It means that not only will large tech corporations have access to advanced tools, but medium-sized companies will also be able to jump on the innovation bandwagon. The future of AI lies in being more accessible and adaptable to the real needs of every business.
