Files
ollama/llm/server.go
Jesse Gross ad6f6a1d29 llm: Change memory allocation backoff from exponential to incremental
If we create a memory layout that should fit based on report free VRAM
but allocation still fails, we start applying a backoff. This reduces
free VRAM by an exponential percentage (1%, 2%, 4%...). However, the
points chosen tend to be too dense at the beginning and too sparse at
the end. Therefore, this switches to an incremental backoff (10%, 20%,
30%...).
2025-10-23 12:58:31 -07:00

53 KiB