Nvidia Keeps China Ties For AI Intelligence Despite Revenue Collapse
——0——
it's still 84 GB of VRAM with 1.57 TB/s pretty much perfect for local inference of whatever that fits. if they have too many, could I have some?
4:15 PM · May 19, 2026 · 3K Views
yeah nah lol but still, arguably a better deal than a Mac Studio
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTexit's still 84 GB of VRAM with 1.57 TB/s pretty much perfect for local inference of whatever that fits. if they have too many, could I have some?
4:15 PM · May 19, 2026 · 3K Views
4:17 PM · May 19, 2026 · 1.6K Views
Like, imagine how gpt-oss-120B flies on this
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)@teortaxesTexyeah nah lol but still, arguably a better deal than a Mac Studio
4:17 PM · May 19, 2026 · 1.6K Views
4:19 PM · May 19, 2026 · 945 Views