7900XT here, previously 6900XT. Support for 6000 series and beyond is great on Linux. You have to use an override envar (Ollama has info on this in their readme). ComfyUI has instructions for anything based on torch.

TensorFlow is unusable.

For the "How many r's in the word strawberry"

    total duration:       15.278476756s
    load duration:        14.982999ms
    prompt eval count:    47 token(s)
    prompt eval duration: 5ms
    prompt eval rate:     9400.00 tokens/s
    eval count:           377 token(s)
    eval duration:        15.257s
    eval rate:            24.71 tokens/s