File	Quant	Size	BPW	Description
`pantheon-ui-lfm25-emoji-merged-Q4_K_M.gguf`	Q4_K_M	~697 MB	4.98	Recommended — best balance of size and quality
`pantheon-ui-lfm25-emoji-merged-Q8_0.gguf`	Q8_0	~1.2 GB	8.50	Near-lossless, best fidelity
`pantheon-ui-lfm25-emoji-merged.f16.gguf`	F16	~2.2 GB	16.00	Full precision, for re-quantization

Usage

A Modelfile with thinking support is included in this repo. Download it alongside the GGUF, then:

bash

ollama create pantheon-emoji -f Modelfile
ollama run pantheon-emoji "your prompt"

To see the model's chain-of-thought reasoning, use the --think flag:

ollama run pantheon-emoji --think "your prompt"

llama-cli -m pantheon-ui-lfm25-emoji-merged-Q4_K_M.gguf -p "Hello" -cnv

Download the GGUF file and load it directly in LM Studio.

Conversion Details

Converted from the original safetensors weights using llama.cpp's convert_hf_to_gguf.py.