Unsloth

A developer benchmarked the Qwen3.6 27B model on Modal using llama.cpp, deploying a serverless pipeline that downloads GGUF shards from Hugging Face and runs perplexity evaluation on an A100-80GB GPU.…

A developer's project to modernize 40-year-old COBOL mainframe code into Python microservices using a fully offline, local AI agent. The developer used the open-source Gemma 4 model, loaded via Unslot…