Modelfile_RTH-LM

# 🦙 Ollama Modelfile for RTH-LM (Fractal TCN)

# 1. Point to the GGUF file we generated
FROM ./rth_lm_25b_v1.gguf

# 2. Set Parameters
PARAMETER temperature 0.7
PARAMETER top_k 40
PARAMETER top_p 0.9

# 3. Define the Prompt Template (Compatible with RTH-LM training)
TEMPLATE """
{{ .Prompt }}
"""

# 4. Deployment Note
# IMPORTANT: This model requires the 'ZetaGrid TCN Patch' for llama.cpp/Ollama 
# to support the Fractal Causal Convolution architecture.
# See: https://huggingface.co/RthItalia/Rth-lm-25b/discussions for setup guide.