This isn't true. You can `ollama run {model}`, `/set parameter num_ctx {ctx}` and then `/save`. Recommended to `/save {model}:{ctx}` to persist on model update
Almost guaranteed this is user error. Ollama has a (tiny) default of 2048 context, so about then is probably when you noticed the results sharply decline in quality. Try 16384
Sure but he explicitly stated, 'GPU Servers', making it likely he didn't use the CPU for inferencing, validating the question about what GPU setup did they use