> Serving at Scale on Chinese AI Chips
> Over the past week, we have served GLM-5.3-Flash on a large-scale cluster of Chinese AI chips, supported by a high-bandwidth interconnect and a serving stack optimized for the underlying hardware.
that's the real story
lain
npub1w8…a786d
2026-08-26 14:57:11 UTC
Author Public Key
npub1w8749su87jassp8h8xjgurl4446c6x8n4qkfdwstvnhzag7mtmusqa786dSeen on
wss://relay.momostr.pinkPublished at
2026-08-26 14:57:11 UTCEvent JSON
{
"id": "adad9658c288ff8ae3c2bbcf825439bcc888fee31ea6ee27680317b1ddffcc75",
"pubkey": "71fd52c387f4bb0804f739a48e0ff5ad758d18f3a82c96ba0b64ee2ea3db5ef9",
"created_at": 1787756231,
"kind": 1,
"tags": [
[
"proxy",
"https://fediffusion.art/objects/9c4b470b-9032-4418-8644-4a786957ea56",
"activitypub"
],
[
"L",
"pink.momostr"
],
[
"l",
"pink.momostr.activitypub:https://fediffusion.art/objects/9c4b470b-9032-4418-8644-4a786957ea56",
"pink.momostr"
],
[
"-"
]
],
"content": "\u003e Serving at Scale on Chinese AI Chips\r\n\u003e Over the past week, we have served GLM-5.3-Flash on a large-scale cluster of Chinese AI chips, supported by a high-bandwidth interconnect and a serving stack optimized for the underlying hardware.\r\n\r\nthat's the real story",
"sig": "e0fb3ad7d7ae244d5d1ed131c2f90655511c1368edfe14fc1391c946eed983f66a6212b81c87e48d30129b1de2e9b348780554c2c95d7e29084015a01f888ca7"
}