Local model
A model that runs on the machine instead of behind an API. Nothing is sent away and there is no per-call cost, which is the trade: privacy and offline use for a smaller, weaker model.
- See also: Quantization
- See also: LLM
- Used in: LLMs
A model that runs on the machine instead of behind an API. Nothing is sent away and there is no per-call cost, which is the trade: privacy and offline use for a smaller, weaker model.