Currently, when deciding which models to run, we are dependent on Groq's inference using Llama-3.3 70b. However, we are currently creating a truly open-source / local model, so it would be nice if we could have a huggingface model or some model that could help in deciding which topics are best for viewing.
Currently, when deciding which models to run, we are dependent on Groq's inference using Llama-3.3 70b. However, we are currently creating a truly open-source / local model, so it would be nice if we could have a huggingface model or some model that could help in deciding which topics are best for viewing.