Downloadable AI models for local, private, customizable, and self-hosted inference.
DeepSeek released V4 Flash Vision’s 305B MIT-licensed weights on Hugging Face. See the timeline, files, benchmarks, local inference, and hardware caveats.
A curated GPT-OSS resource guide for local runtimes, serving, Harmony, coding agents, fine-tuning, implementation verification, and production evaluation.
A practical GPT-OSS guide covering 20B vs 120B, hardware requirements, Ollama and Transformers setup, Harmony, tools, fine-tuning, and AI coding.
