TinyTitan: Streaming Open-Weight LLM Inference from SSD/NVMe Sub-Linear-Memory Mixture-of-Experts Decoding on Apple SiliconBy NewMaxx / November 11, 2026 Direct: https://www.researchgate.net/profile/Andre-Borchert/publication/417559882_TinyTitan_Fast_Streaming_Engine_for_Open-Weight_LLM_Inference_AI_using_SSDNVMe_for_Apple_Silicon_Mac%27s/links/6aede527dc41aa6ac310a722/TinyTitan-Fast-Streaming-Engine-for-Open-Weight-LLM-Inference-AI-using-SSD-NVMe-for-Apple-Silicon-Macs.pdf Reddit: https://www.reddit.com/r/NewMaxx/comments/1xtlwdz/tinytitan_streaming_openweight_llm_inference_from/?utm_source=ifttt