hotepfederales/hotep-kush-v4-gguf

Hotep Intelligence Kush V4 — GGUF

GGUF quantizations of hotepfederales/hotep-kush-v4, the current production model powering askhotep.ai and @hotep_llm_bot.

Available Quantizations

File	Quantization	Size	Use Case
hotep-kush-v4.Q4_K_M.gguf	Q4_K_M	~4.7 GB	Recommended — production default
hotep-kush-v4.Q8_0.gguf	Q8_0	~8.5 GB	Max quality, needs 12+ GB RAM

Model Details

Parameter	Value
Base model	meta-llama/Meta-Llama-3.1-8B-Instruct
LoRA rank	32, alpha 32, RSLoRA
max_seq_length	4096
Quality eval	100/100, 0% CJK, 0% rubric leakage
Persona consistency	10/10

Key Features

• Deep reasoning traces — structured chain-of-thought before every response
• Expanded knowledge base — African history, economics, and philosophy
• Stronger Ma'at alignment — cleaner, culturally grounded answers
• Context retention — handles complex multi-turn conversations

Quick Start (Ollama)

ollama run hotep-llm-kush-v4

Quick Start (llama-cpp-python)

from llama_cpp import Llama

llm = Llama(model_path="hotep-kush-v4.Q4_K_M.gguf", n_ctx=4096)
output = llm("What is the principle of Ma'at?", max_tokens=512)
print(output["choices"][0]["text"])

About Hotep Intelligence

Sovereign AI trained on African history, philosophy, and wisdom traditions — no corporate surveillance, no data collection.

• Website: askhotep.ai
• Telegram bot: @hotep_llm_bot
• Knowledge base: knowledge.askhotep.ai
• Base model (safetensors): hotepfederales/hotep-kush-v4