Evening reflection: "open" Chinese AI models are advancing very rapidly and, as time goes by, they deliver better performance on increasingly limited hardware.
Could the scarcity of RAM and components driven by "traditional" big AI companies be somehow tied to a desire to restrict our access to local computing resources, "forcing" us to pay them for it?
#AI #LLM
Stefano Marinelli
npub14c…j0nj8
2026-08-17 19:12:04 UTC
Author Public Key
npub14calwd6xg349ahf3nnhhyqem2w2e3gs66p7zctz2sna74u3tsddq7j0nj8Seen on
wss://relay.ditto.pubPublished at
2026-08-17 19:12:04 UTCEvent JSON
{
"id": "d7f1914c916e1314e63c9dbc8ea0a24906cd61b4c673fb505b563550ce3604f9",
"pubkey": "ae3bf73746446a5edd319cef72033b539598a21ad07c2c2c4a84fbeaf22b835a",
"created_at": 1786993924,
"kind": 1,
"tags": [
[
"t",
"ai"
],
[
"t",
"llm"
],
[
"proxy",
"https://mastodon.bsd.cafe/users/stefano/statuses/117112433865614057",
"activitypub"
],
[
"client",
"Mostr",
"31990:6be38f8c63df7dbf84db7ec4a6e6fbbd8d19dca3b980efad18585c46f04b26f9:mostr",
"wss://relay.ditto.pub"
]
],
"content": "Evening reflection: \"open\" Chinese AI models are advancing very rapidly and, as time goes by, they deliver better performance on increasingly limited hardware. \nCould the scarcity of RAM and components driven by \"traditional\" big AI companies be somehow tied to a desire to restrict our access to local computing resources, \"forcing\" us to pay them for it?\n\n#AI #LLM",
"sig": "f9ef4cccf911253da1075d040ee4b5c97e36d1bef2e92ef54213939d0f4b992cb9f813a39891a0a2b105956f9935c77d16791d8d3540d2d2cb98e12b49a1ebf5"
}