Intel on X: "New Intel® Arc™ Pro GPUs are on the way. See you in Taipei!"
New Intel® Arc™ Pro GPUs are on the way. See you in Taipei!
Dear Intel, make 32-48GB vram GPU for $600-$900. Local inference is just getting started, you can be leader in this market. Even Apple puts LLMs as main selling point of their Mac Studio historically meant for creative workflows.
If there’s a large vRAM model available, could I borrow one for review? I’ll compare it with my 3090 and ask my university lab colleagues to test it on their workloads—both in cluster and desktop setups—to see if it fits their needs.