prayangshuuu/hummingbird

hummingbird is a lightweight, zero dependency runtime for massive open source Mixture of Experts (MoE) language models. It unifies SSD, RAM, and VRAM into a single intelligent memory hierarchy, enabling inference of models like GPT-OSS 120B, GLM, DeepSeek, Qwen, and more on consumer hardware

语言：C · GitHub 星标：61 · Fork：14

话题：ai-inference · artificial-intelligence · c17 · cross-platform · deepseek · glm · gpt-oss · high-performance-computing

在 GitHub 上查看原仓库