Instructions to use mfielding92/PocketSurvivor-3B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use mfielding92/PocketSurvivor-3B with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("mfielding92/PocketSurvivor-3B", dtype="auto") - llama-cpp-python
How to use mfielding92/PocketSurvivor-3B with llama-cpp-python:
# !pip install llama-cpp-python from llama_cpp import Llama llm = Llama.from_pretrained( repo_id="mfielding92/PocketSurvivor-3B", filename="PocketSurvivor-3B-F16.gguf", )
llm.create_chat_completion( messages = "No input example has been defined for this model task." )
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use mfielding92/PocketSurvivor-3B with llama.cpp:
Install from brew
brew install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama-server -hf mfielding92/PocketSurvivor-3B:Q4_K_M # Run inference directly in the terminal: llama-cli -hf mfielding92/PocketSurvivor-3B:Q4_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama-server -hf mfielding92/PocketSurvivor-3B:Q4_K_M # Run inference directly in the terminal: llama-cli -hf mfielding92/PocketSurvivor-3B:Q4_K_M
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 mfielding92/PocketSurvivor-3B:Q4_K_M # Run inference directly in the terminal: ./llama-cli -hf mfielding92/PocketSurvivor-3B:Q4_K_M
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 mfielding92/PocketSurvivor-3B:Q4_K_M # Run inference directly in the terminal: ./build/bin/llama-cli -hf mfielding92/PocketSurvivor-3B:Q4_K_M
Use Docker
docker model run hf.co/mfielding92/PocketSurvivor-3B:Q4_K_M
- LM Studio
- Jan
- Ollama
How to use mfielding92/PocketSurvivor-3B with Ollama:
ollama run hf.co/mfielding92/PocketSurvivor-3B:Q4_K_M
- Unsloth Studio
How to use mfielding92/PocketSurvivor-3B 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 mfielding92/PocketSurvivor-3B 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 mfielding92/PocketSurvivor-3B to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for mfielding92/PocketSurvivor-3B to start chatting
- Pi
How to use mfielding92/PocketSurvivor-3B with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama-server -hf mfielding92/PocketSurvivor-3B:Q4_K_M
Configure the model in Pi
# Install Pi: npm install -g @mariozechner/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "llama-cpp": { "baseUrl": "http://localhost:8080/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "mfielding92/PocketSurvivor-3B:Q4_K_M" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Hermes Agent new
How to use mfielding92/PocketSurvivor-3B with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama-server -hf mfielding92/PocketSurvivor-3B:Q4_K_M
Configure Hermes
# Install Hermes: curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash hermes setup # Point Hermes at the local server: hermes config set model.provider custom hermes config set model.base_url http://127.0.0.1:8080/v1 hermes config set model.default mfielding92/PocketSurvivor-3B:Q4_K_M
Run Hermes
hermes
- Docker Model Runner
How to use mfielding92/PocketSurvivor-3B with Docker Model Runner:
docker model run hf.co/mfielding92/PocketSurvivor-3B:Q4_K_M
- Lemonade
How to use mfielding92/PocketSurvivor-3B with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull mfielding92/PocketSurvivor-3B:Q4_K_M
Run and chat with the model
lemonade run user.PocketSurvivor-3B-Q4_K_M
List all available models
lemonade list
Install from WinGet (Windows)
winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf mfielding92/PocketSurvivor-3B:# Run inference directly in the terminal:
llama-cli -hf mfielding92/PocketSurvivor-3B: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 mfielding92/PocketSurvivor-3B:# Run inference directly in the terminal:
./llama-cli -hf mfielding92/PocketSurvivor-3B: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 mfielding92/PocketSurvivor-3B:# Run inference directly in the terminal:
./build/bin/llama-cli -hf mfielding92/PocketSurvivor-3B:Use Docker
docker model run hf.co/mfielding92/PocketSurvivor-3B:PocketSurvivor-3B
Model Name: PocketSurvivor-3B
Developed by: mfielding92
License: Apache-2.0
Base Model: unsloth/llama-3.2-3b-instruct-bnb-4bit
Overview
PocketSurvivor-3B is a highly optimized, lightweight language model specifically trained to assist individuals during critical situations such as natural disasters, civil unrest, apocalyptic events, and more. With only 3 billion parameters, this model is small enough to run efficiently on mobile devices, making it ideal for offline deployment in apps like PocketPal.
Key Features
- Action-Oriented Advice: Trained on a synthetic dataset designed to provide clear, actionable steps for survival scenarios, ensuring users can quickly apply the information in life-threatening situations.
- Mental Well-Being Focus: Prioritizes emotional support and mental resilience, offering guidance that helps users remain calm and focused under extreme stress.
- Broad Scenario Coverage: Offers solutions for a variety of scenarios, including food and water procurement, shelter building, self-defense, navigation, and first aid.
- Offline Usability: Optimized to run on mobile devices without requiring internet connectivity, ensuring accessibility in the most challenging circumstances.
- Efficient Training: Built using Unsloth for rapid training, leveraging Hugging Face’s TRL library to maximize efficiency.
Applications
1. Personal Emergency Assistant
PocketSurvivor-3B functions as a multi-faceted survival advisor, offering practical steps for handling events like natural disasters, civil unrest, or large-scale technology failures such as prolonged power or communication outages. It provides guidance on:
- Securing basic needs (food, water, and shelter).
- Delivering first-aid tips.
- Maintaining mental composure under pressure.
While it cannot guarantee safety, it adds an extra layer of support by suggesting clear, actionable strategies when immediate decisions are crucial.
2. Confidence Builder During a Crisis
In the midst of an unfolding crisis, the model helps users adapt to rapidly changing conditions by:
- Providing real-time guidance for resource management and personal safety.
- Assisting with prioritizing tasks and finding temporary solutions to immediate challenges.
- Offering emotional resilience techniques to maintain focus during chaotic situations.
By breaking down overwhelming circumstances into manageable steps, PocketSurvivor-3B empowers users to make informed decisions when it matters most.
3. Resource for Apocalyptic Survival
In the event of large-scale societal collapse or extreme resource scarcity, PocketSurvivor-3B offers practical suggestions for:
- Fortifying living spaces and securing essential resources.
- Optimizing supply usage and planning safe travel routes.
- Maintaining group cohesion, delegating tasks, and managing potential conflicts.
Although it does not replace human judgment or ensure absolute success, it equips users with a range of potential strategies to improve their odds of survival in dire circumstances.
Training Data
The model was trained on a synthetic dataset specifically curated with questions simulating real-world survival scenarios. Data included instructions for:
- Navigating natural disasters like floods, earthquakes, and hurricanes.
- Surviving civil unrest or governmental collapse.
- Managing injuries, illnesses, and stress in isolated conditions.
- Building temporary shelters and purifying water with limited resources.
This dataset emphasizes clarity, adaptability, and prioritization of tasks in high-pressure situations.
Model Performance
PocketSurvivor-3B has been rigorously tested in simulated environments to ensure its responses are practical, efficient, and adaptable to real-world challenges. The model's small size ensures it remains accessible without sacrificing critical functionality.
Disclaimer
While PocketSurvivor-3B is designed to be a helpful companion in critical circumstances, human judgment should always be the final call. Language models, including this one, can hallucinate or provide inaccurate information, especially depending on the level of quantization selected. Users are strongly advised to verify critical information and use the model's outputs as guidance rather than absolute truth. Always exercise caution and common sense in survival situations.
PocketSurvivor-3B empowers users to face adversity with confidence, clarity, and composure, making it a trusted companion in the most unpredictable circumstances.
- Downloads last month
- 400
2-bit
3-bit
4-bit
5-bit
6-bit
8-bit
16-bit

Install from brew
# Start a local OpenAI-compatible server with a web UI: llama-server -hf mfielding92/PocketSurvivor-3B:# Run inference directly in the terminal: llama-cli -hf mfielding92/PocketSurvivor-3B: