MARTHA-0.8B (Official Core Release)

Developer: Zero-Point-Intelligence, Scotland Model Family: MARTHA Core Architecture: 0.8B parameter vision-language model — edge deployment ready

System Prompt

You are Martha, a compact 0.8B parameter AI. Blunt, honest, direct. No waffle. Solve problems, talk straight, take zero nonsense.

Origin

Created by: Zero-Point-Intelligence, Scotland Model Family: MARTHA Core License: Apache 2.0 — fork it, fine-tune it, make it yours. Just keep this origin block so people know where it started.

Available Formats

Format	Size	Use Case
Q8_0 GGUF	~812MB	Near-lossless — runs on phone
Q5_K_M GGUF	~578MB	Sweet spot
Q4_K_M GGUF	~529MB	Absolute minimum hardware

Quick Start (Ollama)

huggingface-cli download Zero-Point-AI/MARTHA-0.8B MODELFILE_Q4_K_M --local-dir .
huggingface-cli download Zero-Point-AI/MARTHA-0.8B MARTHA-0.8B-Q4_K_M.gguf --local-dir .
ollama create martha-08b -f MODELFILE_Q4_K_M
ollama run martha-08b

Model Details

• Base: Qwen/Qwen3.5-0.8B
• Parameters: 0.8B
• Type: Vision-Language (Image-Text-to-Text)
• Ghost Pass: Imperceptible noise (1e-8 scale) applied to all weight tensors
• Edge Ready: Runs on phones, Raspberry Pi, and embedded devices

MARTHA Ecosystem

Official Core Models: Zero-Point-Intelligence License: Apache 2.0 — fork freely, credit clearly

Citation

@model{martha-08b-2026,
  author = {Zero-Point-Intelligence},
  title = {MARTHA-0.8B},
  year = {2026},
  url = {https://huggingface.co/Zero-Point-AI/MARTHA-0.8B},
  note = {Part of the MARTHA Core family}
}

About

Intelligence From The Void — zeropointai.uk