The AMD Instinct MI250X packs 128GB of HBM2e with 3277 GB/s of memory bandwidth, making it one of the fastest cards you can run a local model on. Here is which models fit, how fast they run, and what quantisation gives the best balance of quality and speed.
Already have it? Find the best model for it. Still choosing? See what else $10,000 buys. Or the full specifications.
The AMD Instinct MI250X has 128GB VRAM and 3277 GB/s memory bandwidth, making it one of the fastest cards you can run a local model on. This guide covers which LLMs fit, expected tok/s performance, and recommended settings.
See the full AMD Instinct MI250X specs page for detailed specifications, all compatible models, and speed estimates. Current price: $10000.