GGUF quantizations of
DJLougen/Ornstein-27B — a reasoning-focused fine-tune of Qwen 3.5 27B trained on
1,229 high-quality reasoning traces curated through a custom
Drift Diffusion Modeling (DDM) pipeline.
I'm a PhD student in visual neuroscience at the University of Toronto who also happens to spend way too much time fine-tuning, merging, and quantizing open-weight models on rented H100s and a local DGX Spark. All training compute is self-funded — balancing GPU costs against a student budget. If my uploads have been useful to you, consider buying a PhD student a coffee. It goes a long way toward keeping these experiments running.
Unlike typical reasoning fine-tunes that use large volumes of synthetic data, Ornstein implements quality-over-quantity:
1# Download a quantization (example: Q4_K_M)
2huggingface-cli download DJLougen/Ornstein-27B-GGUF ornstein-27b-q4_k_m.gguf --local-dir .
3
4# Run with llama.cpp
5./llama-cli -m ornstein-27b-q4_k_m.gguf \
6 -p "You are a helpful reasoning assistant." \
7 --temp 0.6 -n 8192
1# Create a Modelfile
2cat <<EOF > Modelfile
3FROM ./ornstein-27b-q4_k_m.gguf
4PARAMETER temperature 0.6
5PARAMETER num_predict 8192
6SYSTEM "You are a helpful reasoning assistant."
7EOF
8
9ollama create ornstein -f Modelfile
10ollama run ornstein
1@misc{ornstein27b,
2 author = {DJLougen},
3 title = {Ornstein-27B: DDM-Curated Reasoning Fine-Tune of Qwen 3.5 27B},
4 year = {2026},
5 publisher = {Hugging Face},
6 url = {https://huggingface.co/DJLougen/Ornstein-27B}
7}
