Erik S

MacOS: Ollama ignores changes to the iogpu.wired_limit_mb tunable when deciding whether to run on GPU or CPU

13

MacOS 14.2.1 on a 32GB M1 Max MBP ``` % ollama run dolphin-mixtral:8x7b-v2.7-q3_K_M Error: model requires at least 48 GB of memory ``` This error appears immediately, it does not...

On MacOS, I can't select a symlink to a local model GGUF

6

I use Ollama and have a directory of symlinks with human-readable names to the actual GGUF blobs of models I've downloaded into my local ollama repository. When I try out...

People are often concerned about what they perceive to be low performance and/or whether Ollama is making optimal use of their RAM/VRAM/GPU/CPU cores. This comes up frequently in github issues,...

Add resistance when reaching the edge of another window

1

When using the native MacOS functionality to arrange & resize windows there is some "resistance" when the edge of the window reaches the parallel or co-linear edge of an adjacent...