History
The team behind llama.cpp joins Hugging Face
On 20 February 2026, ggml.ai announced it was joining Hugging Face to support the further development of local AI.

Software on your own computer also needs people who keep maintaining it. On 20 February 2026, ggml.ai announced it was joining Hugging Face. This is the founding team behind llama.cpp. The announcement concerned the future of local AI: using models on hardware you choose yourself.

The source says the team will continue to lead and maintain the projects. It also describes support for the community around those projects. The news was therefore mainly a decision about lasting maintenance. It was not the announcement of one new model that you could immediately use for every task.

For a small business, this brings a different question into view. Who keeps the software working when you run a model on your own equipment? Consider that alongside the quality of the answers. In a trial, ask who installs updates, what equipment is needed and how you return to an earlier setup if a change disappoints.
Start with public manuals, for example, so you can assess the technology without using business data. First check whether the result is useful. Want to understand which parts you need to manage yourself? Discuss your setup with Denkbaar.