Comment

Comment by 5kg

Host to device bandwidth (ram to vram) is 128Gb/s for PCIe Gen 6. VRAM to GPU bandwidth is 1.8Tb/s for GDDR 7 (5090), and 8Tb/s for HBM3e (B200). So it can be faster to recompute than offload kv cache.
parent 5kg 5 months ago

No comments yet

Be the first to share your thoughts.

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton