The problem with using closed-source large models like OpenAI's ChatGPT is that your data could potentially be used as training material for the model at any time.
While open-source large models are now capable of matching the performance of closed-source large models, and many platforms also provide LLM inference services based on these open-source models, users of these services still have to bear the same data privacy risks, as these platforms may also collect user information to train their large models.
The best way to protect data privacy and security at the moment is to run LLMs on the user's own device. All the computation is carried out on the user's personal device, and the user's data never leaves their own device, so there will be no risk of data leaks.
However, for users to run LLM models on their own devices, they need to go through a series of complex setup processes and have some understanding of the principles behind LLM models. For example:
In short, you need to be someone with an interest in technology in order to properly set up the local running of an LLM.
The mission of FishLLM is to make it possible for as many people as possible to use LLM models on their own devices.
FishLLM helps you handle all the complex setup, you don't need to understand the principles of LLM, and you don't need to be tech-savvy either. You just need to click to install, and FishLLM will complete all the setup for you.
FishLLM will track the latest open-source LLM models, and based on your device's hardware configuration, it will find the model that is most suitable for your device. You only need to click the update button periodically to use the latest open-source models, without having to waste time searching for latest models.