Kindi 0.4: local LLM inference on low-cost NVIDIA GPUs, Radeon 780M and CPUs. Bonsai 2B-27B, AH-Nanbeige4.3-33B and experimental Nemotron, with an OpenAI-compatible server.
↑ +1 today
Discover the best Maxwell open source repositories on GitHub. Explore trending tools, live star counts, source code, and developer projects updated daily.