Deepseek-R1 not doing reasoning, llama.cpp, "MemoryError"

Forgive me, I am very new to AI and just coming into it with some light programming and decent linux knowledge. I am running Deepseek-R1-Distill on llama.cpp, and when using the webui, it does not go through its reasoning process prior to answering. When I use it from the command line using llama-cli -m, it thinks before it returns an answer.

Also, I am getting the following error for some models

`build: 4767 (08d59862) with cc (Gentoo Hardened 13.3.1_p20241025 p1) 13.3.1 20241024 for x86_64-pc-linux-gnu main: llama backend init main: load the model and apply lora adapter, if any gguf_init_from_file_impl: invalid magic characters: '<!do', expected 'GGUF' llama_model_load: error loading model: llama_model_loader: failed to load model from zephyr-7b-beta.Q8_0.gguf

llama_model_load_from_file_impl: failed to load model common_init_from_params: failed to load model 'zephyr-7b-beta.Q8_0.gguf' main: error: unable to load model `

Not sure if it's an issue with the way I have it configured, or what. Help is appreciated <3