Fine-tuning takes an already-trained general model and continues training it on a smaller, task-specific dataset so it better matches a desired style, tone, or task. Methods like LoRA update only a small set of parameters, keeping cost and barriers low.
Enclave connects to models via BYOK and does not require fine-tuning — most personalization comes from personas, long-term memory, and prompts. If you have a self-trained or fine-tuned model, you can plug it in via a local model or an OpenAI-compatible endpoint.
Try it yourself in Enclave
Open it in your browser — no credit card, no install.