Crawling the web models that run on 16 GB
2 models with published weights that fit in 16 GB — the common laptop. Counted at four-bit quantisation, weights only, leaving the machine about a third of its memory and a gigabyte for context: room for roughly 16 billion parameters. A mixture of experts is counted in full, because it is held in full even though only a few experts compute.
The commonest laptop configuration, and the point where a genuinely useful model runs locally: fourteen billion parameters at four-bit fits with room to spare for the context. Expect a capable general assistant that will not match the frontier, and remember that anything else the machine is doing competes for the same memory.
Tools that fetch pages and hand them back as something a model can read: rendered HTML, markdown, a screenshot, a whole site map. Priced per page or per session. The differences that matter are whether JavaScript is executed, what happens at a login or a bot check, and how gracefully the thing fails on a site that does not want to be read.
- xlm-roberta-largeFacebook AI community0.6B≈0.4 GB at 4-bit
- xlm-roberta-baseFacebook AI community0.3B≈0.2 GB at 4-bit