https://www.datatalkwithankit.com LLM inference benchmarks, GPUs, serving configs, and cost, by Ankit Dongare. en-us Sun, 11 Oct 2026 12:00:00 GMT https://www.datatalkwithankit.com/posts/h200-vs-mi300x https://www.datatalkwithankit.com/posts/h200-vs-mi300x Memory, interconnect, and price compared across NVIDIA H200 and AMD MI300X: what the extra 51 GB buys, why PCIe vs NVLink decides multi-GPU scaling, and what a GPU-hour really costs on the neoclouds. Fri, 21 Aug 2026 09:00:00 GMT