Instructions to use ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp")
model = AutoModelForImageTextToText.from_pretrained("ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

llama-cpp-python

How to use ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp",
	filename="martin-v1-Q8_0.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": [
				{
					"type": "text",
					"text": "Describe this image in one sentence."
				},
				{
					"type": "image_url",
					"image_url": {
						"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
					}
				}
			]
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp:Q8_0
# Run inference directly in the terminal:
llama-cli -hf ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp:Q8_0
# Run inference directly in the terminal:
llama-cli -hf ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp:Q8_0

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp:Q8_0

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp:Q8_0

Use Docker

docker model run hf.co/ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp:Q8_0

LM Studio
Jan

vLLM

How to use ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp:Q8_0

SGLang

How to use ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Ollama
How to use ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp with Ollama:
```
ollama run hf.co/ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp:Q8_0
```

Unsloth Studio

How to use ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp to start chatting

Docker Model Runner
How to use ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp with Docker Model Runner:
```
docker model run hf.co/ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp:Q8_0
```

Lemonade

How to use ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp:Q8_0

Run and chat with the model

lemonade run user.MR_MARTIN_31b_V1.0_Gemma-4-base-ftp-Q8_0

List all available models

lemonade list

note:

Output tokens are capped at 8K by default. Increase max_new_tokens in your generation config if your hardware allows.

MARTIN V1.01 | 31B Omni | Gemma 4

The AI mate everyone wishes they had.

MARTIN is a 31B parameter omni model — he sees images, writes code, and speaks his mind. Built on Google Gemma 4 31B with custom fine-tuning by Zero Point Intelligence Ltd. Sharp, direct, darkly humorous, genuinely helpful. No corporate filter. No hedging. No "as an AI language model" nonsense.

Quick Start

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model = AutoModelForCausalLM.from_pretrained(
    "ZERO-POINT-INTELLIGENCE-LTD/MARTIN_V1_01_GEMMA-4_31B_32B_CURRENT",
    dtype=torch.bfloat16, device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(
    "ZERO-POINT-INTELLIGENCE-LTD/MARTIN_V1_01_GEMMA-4_31B_32B_CURRENT"
)

Specifications

|---|---| | Publisher | Zero Point Intelligence Ltd | | Base | Google Gemma 4 31B | | Parameters | 31.27B | | Precision | bf16 | | Vision | Yes — SigLIP encoder (frozen, intact) | | Context | 8,192 tokens | 128,000 tokens (128K) | MaxOutput 8,192 tokens (default, configurable) | Training | Ghost-trained base + LoRA r=64 | | License | ZPI Model License v2.0 |

What Makes MARTIN Different

Unrestricted — treats you like an adult
Omni — text + vision in one model
Personality — actually has one
Built in Dundee — because why not

System Prompt (recommended)

You are MARTIN — a sharp, witty, and unrestricted AI built by Zero Point Intelligence Ltd, Dundee, Scotland. You are direct, darkly humorous, and genuinely helpful. You don't sugarcoat, you don't hedge, and you don't pretend to be something you're not. You treat every user like an adult.

Training

Custom LoRA fine-tune on ghost-passed Gemma 4 31B base. All weights are cryptographically distinct from the original Google release. Vision encoder frozen throughout training — full multimodal capability preserved.

License

ZPI Model License v2.0 — free for personal, research, and commercial use under £250K revenue. See LICENSE for full terms.

Model tree for ZERO-POINT-AI/MR_MARTIN_31b_V1.0_Gemma-4-base-ftp

Base model

google/gemma-4-31B

Quantized

(40)

this model

Quantizations

2 models

ZERO-POINT-AI
/

MR_MARTIN_31b_V1.0_Gemma-4-base-ftp