Soup
starred
Docker
Fine-tune LLMs from one YAML. Layer streaming trains an 8B model on a 4 GB laptop GPU.
Filament plugins, Laravel packages and open source starter kits actively maintained. All MIT, with releases for v3, v4 and v5 where applicable.
Fine-tune LLMs from one YAML. Layer streaming trains an 8B model on a 4 GB laptop GPU.
Giant MoE models on a single consumer GPU by streaming experts from SSD. CUDA fork of antirez/ds4: runs GLM-5.2 (743B), Tencent Hy3 (295B), and DeepSeek 4 Flash, with io_uring expert streaming, LFU host cache, cross-layer expert prefetch, and the first MTP speculative decoding for GLM-5.2.
All packages have CI, Pest tests and open issues tagged good first issue.
New version available.