fngarrett··on Aiter: AI Tensor Engine for ROCmWe have a dual W7800 system in-house as our `gfx1100` rig. I'll try to install and run through the tests sometime this week.
fngarrett··on Exploring inference memory saturation effect: H100 vs. MI300xGreat read. Have you compared performance with other Llama models (3, 3.2) or have you just done benchmarking with 3.1?Is there some intuition as to why 3.1 might outperform 3.2 on MI300X?
fngarrett··on Eighty Years of the Finite Element Method (2022)For anyone interested in a contemporary implementation, SELF is a spectral element library in object-oriented fortran [1]. The devs here at Fluid Numerics have upcoming benchmarks on our MI300A system and other cool hardware.[1] https://github.com/FluidNumerics/SELF