Instructions to use ntsd/zerocoach-athletic-qwen3-1.7b-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use ntsd/zerocoach-athletic-qwen3-1.7b-GGUF with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf ntsd/zerocoach-athletic-qwen3-1.7b-GGUF:Q4_K_M # Run inference directly in the terminal: llama cli -hf ntsd/zerocoach-athletic-qwen3-1.7b-GGUF:Q4_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf ntsd/zerocoach-athletic-qwen3-1.7b-GGUF:Q4_K_M # Run inference directly in the terminal: llama cli -hf ntsd/zerocoach-athletic-qwen3-1.7b-GGUF:Q4_K_M
Build from source code
git clone https://github.com/ggml-org/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-app # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama serve -hf ntsd/zerocoach-athletic-qwen3-1.7b-GGUF:Q4_K_M # Run inference directly in the terminal: ./build/bin/llama cli -hf ntsd/zerocoach-athletic-qwen3-1.7b-GGUF:Q4_K_M
Use Docker
docker model run hf.co/ntsd/zerocoach-athletic-qwen3-1.7b-GGUF:Q4_K_M
- Llama new
- LM Studio
- Jan
- vLLM
How to use ntsd/zerocoach-athletic-qwen3-1.7b-GGUF with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "ntsd/zerocoach-athletic-qwen3-1.7b-GGUF" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "ntsd/zerocoach-athletic-qwen3-1.7b-GGUF", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/ntsd/zerocoach-athletic-qwen3-1.7b-GGUF:Q4_K_M
- Ollama
How to use ntsd/zerocoach-athletic-qwen3-1.7b-GGUF with Ollama:
ollama run hf.co/ntsd/zerocoach-athletic-qwen3-1.7b-GGUF:Q4_K_M
- Unsloth Desktop
- Pi
How to use ntsd/zerocoach-athletic-qwen3-1.7b-GGUF with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf ntsd/zerocoach-athletic-qwen3-1.7b-GGUF:Q4_K_M
Configure the model in Pi
# Install Pi: npm install -g @earendil-works/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "llama-cpp": { "baseUrl": "http://localhost:9931/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "ntsd/zerocoach-athletic-qwen3-1.7b-GGUF:Q4_K_M" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use ntsd/zerocoach-athletic-qwen3-1.7b-GGUF with Docker Model Runner:
docker model run hf.co/ntsd/zerocoach-athletic-qwen3-1.7b-GGUF:Q4_K_M
- Lemonade
How to use ntsd/zerocoach-athletic-qwen3-1.7b-GGUF with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull ntsd/zerocoach-athletic-qwen3-1.7b-GGUF:Q4_K_M
Run and chat with the model
lemonade run user.zerocoach-athletic-qwen3-1.7b-GGUF-Q4_K_M
List all available models
lemonade list
- Hermes Agent
How to use ntsd/zerocoach-athletic-qwen3-1.7b-GGUF with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf ntsd/zerocoach-athletic-qwen3-1.7b-GGUF: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://127.0.0.1:9931/v1 hermes config set model.default ntsd/zerocoach-athletic-qwen3-1.7b-GGUF:Q4_K_M
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use ntsd/zerocoach-athletic-qwen3-1.7b-GGUF with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf ntsd/zerocoach-athletic-qwen3-1.7b-GGUF:Q4_K_M
Configure OpenClaw
# Install OpenClaw: npm install -g openclaw@latest # Register the local server and set it as the default model: openclaw onboard --non-interactive --mode local \ --auth-choice custom-api-key \ --custom-base-url http://127.0.0.1:9931/v1 \ --custom-model-id "ntsd/zerocoach-athletic-qwen3-1.7b-GGUF:Q4_K_M" \ --custom-provider-id llama-cpp \ --custom-compatibility openai \ --custom-text-input \ --accept-risk \ --skip-health
Run OpenClaw
openclaw agent --local --agent main --message "Hello from Hugging Face"
ZeroCoach Athletic Intelligence v1 (Qwen 3 1.7B - GGUF)
ZeroCoach Athletic is an elite, sovereign on-device athletic intelligence and sports science language model fine-tuned on top of Qwen 3 1.7B. It is designed to run 100% locally on smartphones (iOS via Metal / Android via OpenCL) without cloud subscriptions, keeping biometric telemetry completely private.
๐ Key Features
- Physiological Telemetry Autoregulation: Dynamically calculates workout intensity, volume caps, and deload protocols based on real-time biometric readiness scores, HRV, and sleep history.
- Biomechanical Precision: Incorporates strict kinetic chain cues, bar paths, scapular mechanics, and calisthenics leverage progression trees.
- Precision Nutrition & Leucine Synthesis: Calculates exact gram-level macronutrient allocations (Protein, Carbs, Fat, Fiber) aligned with active calorie expenditure.
- Mobile-Engineered GGUF: Quantized with
Q4_K_Mto 1.05 GB, streaming at ~63 tokens/second with low battery consumption and sub-1.2 GB active RAM footprint.
๐ Fine-Tuning Benchmarks & Training Telemetry
- Base Model:
Qwen/Qwen3-1.7B - Architecture: Causal LM with Native ChatML (
<|im_start|>/<|im_end|>) - Hardware: Trained with Apple Silicon / NVIDIA Blackwell Tensor Cores (
bfloat16) - Trainable Parameters: 17,432,576 (adapted across all attention & MLP projections)
- Final Validation Loss: 0.0588
- Context Length: 1,024 โ 2,048 tokens
๐ฆ Quantization Specifications
| Quantization | File Size | Bits Per Weight | Memory Footprint | Recommended Device |
|---|---|---|---|---|
| Q4_K_M | 1.05 GB (1,050 MiB) | 5.12 BPW | ~1.1 GB RAM | Modern Smartphones (iPhone 12+, 4GB+ Androids) |
๐ ๏ธ Usage with Ollama
ollama run hf.co/ntsd/zerocoach-athletic-qwen3-1.7b-GGUF
๐ฑ Usage in ZeroCoach Mobile App
In the ZeroCoach app:
- Open Settings -> Model Manager.
- Select ZeroCoach Athletic (Qwen 3 1.7B Fine-Tuned).
- Tap Download Model to load weights into local phone storage.
๐ License
Apache 2.0
- Downloads last month
- 140
4-bit