@[email protected] to [email protected]English • 1 year agoSnowden: "They've gone full mask-off: do not ever trust OpenAI or its products"twitter.commessage-square177fedilinkarrow-up1626cross-posted to: [email protected]
arrow-up1626external-linkSnowden: "They've gone full mask-off: do not ever trust OpenAI or its products"twitter.com@[email protected] to [email protected]English • 1 year agomessage-square177fedilinkcross-posted to: [email protected]
minus-squareclassiclinkfedilink14•1 year agoIs there a magazine or site that breaks this down for the less tech savvy? And is the quality of the AI on par?
minus-square@[email protected]linkfedilink9•1 year agoOn par? No. Good enough? Definitely. Ollama baby
minus-square@[email protected]linkfedilink21•1 year agoCheck my notes https://fabien.benetou.fr/Content/SelfHostingArtificialIntelligence but as others suggested a good way to start is probably https://github.com/ollama/ollama/ and if you need a GUI https://gpt4all.io
minus-square@[email protected]linkfedilink5•1 year agoI’m not the person who asked, but still thanks for the information. I might give this a try soon.
minus-square@[email protected]linkfedilink1•1 year ago You should have at least 16 GB of RAM available to run the 13B models, Is this gpu ram or cpu ram?
minus-square@[email protected]linkfedilinkEnglish1•1 year agoEither works, but system RAM is at least an order of magnitude slower, more play by mail than chat…
minus-squareKillingTimeItselflinkfedilinkEnglish2•1 year agolikely GPU ram, there is some tech that can offload ram, but generally it’s all hosted in VRAM, this requirement will likely fade as NPUs start becoming a thing though.
minus-square@[email protected]linkfedilink1•1 year agopretty sure it can run on either, but cpus are slow compared to gpus, often to the point of being impractical
Is there a magazine or site that breaks this down for the less tech savvy? And is the quality of the AI on par?
Your best bet is YouTubing ollama.
Ollama with Lava and Mistral
On par? No. Good enough? Definitely. Ollama baby
Check my notes https://fabien.benetou.fr/Content/SelfHostingArtificialIntelligence but as others suggested a good way to start is probably https://github.com/ollama/ollama/ and if you need a GUI https://gpt4all.io
I’m not the person who asked, but still thanks for the information. I might give this a try soon.
Ditto, thanks to everyone’s for their suggestions
Is this gpu ram or cpu ram?
Either works, but system RAM is at least an order of magnitude slower, more play by mail than chat…
likely GPU ram, there is some tech that can offload ram, but generally it’s all hosted in VRAM, this requirement will likely fade as NPUs start becoming a thing though.
pretty sure it can run on either, but cpus are slow compared to gpus, often to the point of being impractical