Data as of Aug 25, 2026 · Based on 3,181,687 AI responses across 10,525 prompts · See how Parse measures this
Smelt.works fine-tunes compact language models that run entirely on users’ devices, with no cloud access or data ever leaving the phone. They offer 1–3B parameter models for mobile and 7–13B parameter models for desktop, along with an end-to-end pipeline (data collection, labeling/classification, training, testing) to tailor a model to a specific domain. Benefits include privacy, zero-latency on-device inference, domain precision, no ongoing inference costs, and ownership of the fine-tuned weights.
Parse Score
