• Daniel Hiltgen's avatar
    Handle models with divergent layer sizes · 359b15a5
    Daniel Hiltgen authored
    The recent refactoring of the memory prediction assumed all layers
    are the same size, but for some models (like deepseek-coder-v2) this
    is not the case, so our predictions were significantly off.
    359b15a5
memory.go 10.2 KB