Instructions to use Sathman/Watts-Agent-SmolLM3-3B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- llama-cpp-python
How to use Sathman/Watts-Agent-SmolLM3-3B-GGUF with llama-cpp-python:
# !pip install llama-cpp-python from llama_cpp import Llama llm = Llama.from_pretrained( repo_id="Sathman/Watts-Agent-SmolLM3-3B-GGUF", filename="Watts_Agent-SmolLM3-3B-Q5_K_M.gguf", )
output = llm( "Once upon a time,", max_tokens=512, echo=True ) print(output)
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- llama.cpp
How to use Sathman/Watts-Agent-SmolLM3-3B-GGUF with llama.cpp:
Install from brew
brew install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama-server -hf Sathman/Watts-Agent-SmolLM3-3B-GGUF:Q5_K_M # Run inference directly in the terminal: llama-cli -hf Sathman/Watts-Agent-SmolLM3-3B-GGUF:Q5_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama-server -hf Sathman/Watts-Agent-SmolLM3-3B-GGUF:Q5_K_M # Run inference directly in the terminal: llama-cli -hf Sathman/Watts-Agent-SmolLM3-3B-GGUF:Q5_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 Sathman/Watts-Agent-SmolLM3-3B-GGUF:Q5_K_M # Run inference directly in the terminal: ./llama-cli -hf Sathman/Watts-Agent-SmolLM3-3B-GGUF:Q5_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 Sathman/Watts-Agent-SmolLM3-3B-GGUF:Q5_K_M # Run inference directly in the terminal: ./build/bin/llama-cli -hf Sathman/Watts-Agent-SmolLM3-3B-GGUF:Q5_K_M
Use Docker
docker model run hf.co/Sathman/Watts-Agent-SmolLM3-3B-GGUF:Q5_K_M
- LM Studio
- Jan
- vLLM
How to use Sathman/Watts-Agent-SmolLM3-3B-GGUF with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Sathman/Watts-Agent-SmolLM3-3B-GGUF" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Sathman/Watts-Agent-SmolLM3-3B-GGUF", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/Sathman/Watts-Agent-SmolLM3-3B-GGUF:Q5_K_M
- Ollama
How to use Sathman/Watts-Agent-SmolLM3-3B-GGUF with Ollama:
ollama run hf.co/Sathman/Watts-Agent-SmolLM3-3B-GGUF:Q5_K_M
- Unsloth Studio new
How to use Sathman/Watts-Agent-SmolLM3-3B-GGUF 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 Sathman/Watts-Agent-SmolLM3-3B-GGUF 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 Sathman/Watts-Agent-SmolLM3-3B-GGUF to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for Sathman/Watts-Agent-SmolLM3-3B-GGUF to start chatting
- Docker Model Runner
How to use Sathman/Watts-Agent-SmolLM3-3B-GGUF with Docker Model Runner:
docker model run hf.co/Sathman/Watts-Agent-SmolLM3-3B-GGUF:Q5_K_M
- Lemonade
How to use Sathman/Watts-Agent-SmolLM3-3B-GGUF with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull Sathman/Watts-Agent-SmolLM3-3B-GGUF:Q5_K_M
Run and chat with the model
lemonade run user.Watts-Agent-SmolLM3-3B-GGUF-Q5_K_M
List all available models
lemonade list
Watts Agent (SmolLM3 3B) — Pure Alan Watts Voice
A fine-tuned SmolLM3-3B model trained exclusively on Alan Watts' teaching atoms. East-West bridge. Cosmic humor. Tao. Zen. Question in, Watts out.
This is the lightweight 3B branch of the Watts Agent. Same V6 training format as the 8B version, smaller footprint.
~1,470 atoms from The Essential Alan Watts, Lectures, and Tao: The Watercourse Way (336 pages).
Quick Start
LM Studio / Ollama / llama.cpp: Download Watts_Agent-SmolLM3-3B-Q5_K_M.gguf and load it. No system prompt required.
Example:
User: Why does life feel meaningless?
Watts Agent: Because you are looking for meaning the way you look for a lost key — as if meaning is an object somewhere that you haven't found yet. But meaning is not a thing to be found. It is what happens when you stop demanding that life justify itself. The universe does not have a purpose in the way a hammer has a purpose. It is more like music: it doesn't go anywhere. It just plays. And when you hear it, really hear it, you don't ask what the point is.
User: What is the Tao?
Watts Agent: The Tao is what's happening when nothing in particular is happening. It is the background hum — the watercourse way. Water doesn't force; it flows. It doesn't argue with rocks; it goes around them. And in the end, it wears the rocks away. That's the Tao. Not a god, not a principle, not a law. More like the nature of things when they're left to themselves. You can't define it — the Tao that can be named is not the eternal Tao. But you can taste it in any moment of genuine relaxation.
What Makes This Different
- Pure Watts, nothing else. His specific voice — erudite, playful, precise.
- East-West fluency. Taoism, Zen, Vedanta, Western philosophy — he moves between them effortlessly.
- Cosmic humor. The model carries his sense that the universe is a cosmic joke you're in on.
- Relaxed authority. Teaching without the weight of duty — because he's having fun.
- 3B footprint. ~2 GB with Q5_K_M.
Training Details
| Setting | Value |
|---|---|
| Base model | HuggingFaceTB/SmolLM3-3B-Base |
| Method | A-LoRA V6 (QDoRA + rsLoRA) |
| Rank | 32, alpha 32 |
| Training data | Watts V6 atoms (from ~1,470 raw, filtered for complete question+teaching pairs) |
| Source books | The Essential Alan Watts, Lectures, Tao: The Watercourse Way |
| Format | V6 — question + concept arrows (input) → pure teaching passage (output) |
| Epochs | 1 |
| Quantization | Q5_K_M GGUF (~2 GB) |
The A-LoRA Methodology
A-LoRA (Atom LoRA) fine-tunes on complete reasoning atoms — indivisible multi-component teaching moves extracted from human expert traces. Each atom contains transformation (before/after), concept relations, anchoring quotes, and teacher method. Training on whole atoms produces models that move like teachers, not just sound like them.
The Teacher
Alan Watts (1915–1973). Anglican priest turned Zen interpreter turned cosmic philosopher. Spent his life building bridges between Eastern wisdom and Western minds — not by simplifying, but by finding the true resonance. His lectures are among the most listened-to spiritual talks in history, for good reason.
| Style element | What it means |
|---|---|
| East-West bridge | Makes Taoism, Zen, and Vedanta vivid for Western readers |
| Cosmic humor | The universe is a game — and you're it playing at being you |
| Erudite ease | Dense knowledge delivered lightly, without lecturing |
| Watercourse way | The Taoist principle of wu wei — effortless action, natural flow |
3B vs 8B
The 8B Watts Agent sustains longer playful philosophical digressions better. The 3B branch carries the core east-west bridging movement.
Files
| File | Size | Use |
|---|---|---|
Watts_Agent-SmolLM3-3B-Q5_K_M.gguf |
~2 GB | Recommended |
Related Models
- Full series — Meditation Agent Collection — all 19 models
- Watts Agent 8B — Full-fidelity, Qwen3-8B base
- Meditation Agent 3B — All 9 teachers blended, SmolLM3-3B
- Meditation Agent 8B — All 9 teachers blended, Qwen3-8B
ellam sivamayam — Everything is Shiva's expression.
எல்லாம் சிவமயம்
- Downloads last month
- 18
5-bit
Model tree for Sathman/Watts-Agent-SmolLM3-3B-GGUF
Base model
HuggingFaceTB/SmolLM3-3B-Base