Mini Trainer implements Orthogonal Subspace Fine-Tuning (OSFT), a breakthrough continual learning technique that enables models to learn new tasks without catastrophic forgetting. OSFT uses adaptive ...
Kimi-K2-Mini is an experimental compressed version of the 1.07T parameter Kimi-K2 model, targeting ~32.5B parameters for more accessible deployment. This project explores several optimization ...
Given what we’ve discussed on AppStories lately, it shouldn’t come as a huge surprise that I’ve been using Claude Code on my Mac mini home server a lot. And if you know me, you shouldn’t be surprised ...
Abstract: Grid-connected inverters heavily depend on the grade of current detection and processing devices since their performance is assessed by the quality of the controlled currents. However, ...