{"type":"rich","version":"1.0","provider_name":"Transistor","provider_url":"https://transistor.fm","author_name":"Machine Learning Tech Brief By HackerNoon","title":"WiCi Infer Runs 800B-Parameter AI Models on a Laptop by Streaming Them From the SSD","html":"<iframe width=\"100%\" height=\"180\" frameborder=\"no\" scrolling=\"no\" seamless src=\"https://share.transistor.fm/e/55d745ba\"></iframe>","width":"100%","height":180,"duration":375,"description":"\n        This story was originally published on HackerNoon at: https://hackernoon.com/wici-infer-runs-800b-parameter-ai-models-on-a-laptop-by-streaming-them-from-the-ssd.\nWiCi Infer is a free desktop app that runs 800B-parameter AI models on Macs and PCs by streaming weights from the SSD. Here is how it works and how fast it is.\nCheck more stories related to machine-learning at: https://hackernoon.com/c/machine-learning.\n            You can also check exclusive content about #artificial-intelligence, #machine-learning, #local-ai, #llm, #business, #consumer-hardware-ai, #gpu, #open-source,  and more.\nThis story was written by: @sarahevans. Learn more about this writer by checking @sarahevans's about page,\n            and for more stories, please visit hackernoon.com.\nWiCi Infer is a free desktop app for Mac and Windows that runs 800B-parameter open-weight models, including DeepSeek V4.1 Flash and Meta's Muse Spark Open, on computers with as little as 16GB of memory. It keeps the weights on the SSD and loads only what each step needs. WiCi's benchmarks show about 2.7 tokens per second on a 48GB MacBook Pro. The cost: roughly 520GB of free storage and modest speed.","thumbnail_url":"https://img.transistorcdn.com/KyA01h2FD2insgk-wX_xzV6vbJnTNl2BvPYVL-XaI9A/rs:fill:0:0:1/w:400/h:400/q:60/mb:500000/aHR0cHM6Ly9pbWct/dXBsb2FkLXByb2R1/Y3Rpb24udHJhbnNp/c3Rvci5mbS9zaG93/LzQxMjcyLzE2ODM1/ODI0ODgtYXJ0d29y/ay5qcGc.webp","thumbnail_width":300,"thumbnail_height":300}