license: mit
base_model: OpenIntelligenceNet/Phi-4-mini-instruct-Abliterated
library_name: gguf
pipeline_tag: text-generation
language:
- en
tags: - gguf
- llama.cpp
- phi-4
- phi-4-mini
- abliterated
- uncensored
- imatrix
- quantized
Phi-4-mini-instruct-Abliterated-GGUF
GGUF quantizations of OpenIntelligenceNet/Phi-4-mini-instruct-Abliterated,
which is an abliterated version of microsoft/Phi-4-mini-instruct.
All quants were made with llama.cpp using an importance matrix (imatrix.dat, included).
| File | Quant | Notes |
|---|---|---|
| phi4-mini-Q3_K_M.gguf | Q3_K_M | Smallest, lowest quality |
| phi4-mini-Q4_K_M.gguf | Q4_K_M | Good size/quality balance |
| phi4-mini-Q5_K_M.gguf | Q5_K_M | Higher quality |
| phi4-mini-Q6_K.gguf | Q6_K | Near-lossless |
| phi4-mini-Q8_0.gguf | Q8_0 | Highest quality quant |
Usage
llama-cli -m phi4-mini-Q4_K_M.gguf -p "Hello" -cnv
Limitations and responsible use
This model has reduced refusal behavior and may produce harmful, inaccurate, or offensive content. It has not
undergone formal safety or benchmark evaluation. You are responsible for how you use it and for complying with
applicable laws and the base model's MIT license. Do not use it to facilitate violence, weapons, fraud, or harm to others.