Crawling the web models that run on 8 GB
2 models with published weights that fit in 8 GB — a phone, a base iPad, an Air. Counted at four-bit quantisation, weights only, leaving the machine about a third of its memory and a gigabyte for context: room for roughly 7 billion parameters. A mixture of experts is counted in full, because it is held in full even though only a few experts compute.
The memory of a phone, a base iPad or an entry-level laptop. What fits is small: models of a few billion parameters, quick and cheap to run, good at summarising, classifying and simple extraction, and out of their depth on long reasoning. This is also where on-device makes the most sense, because the alternative is a network round trip for something that takes a moment.
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