Instructions to use Sathman/Tolle-Agent-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Sathman/Tolle-Agent-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Sathman/Tolle-Agent-GGUF",
	filename="Tolle_Agent-Q8_0.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use Sathman/Tolle-Agent-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/Tolle-Agent-GGUF:Q8_0
# Run inference directly in the terminal:
llama-cli -hf Sathman/Tolle-Agent-GGUF:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Sathman/Tolle-Agent-GGUF:Q8_0
# Run inference directly in the terminal:
llama-cli -hf Sathman/Tolle-Agent-GGUF: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 Sathman/Tolle-Agent-GGUF:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf Sathman/Tolle-Agent-GGUF: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 Sathman/Tolle-Agent-GGUF:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Sathman/Tolle-Agent-GGUF:Q8_0

Use Docker

docker model run hf.co/Sathman/Tolle-Agent-GGUF:Q8_0

LM Studio
Jan

vLLM

How to use Sathman/Tolle-Agent-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Sathman/Tolle-Agent-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": "Sathman/Tolle-Agent-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Sathman/Tolle-Agent-GGUF:Q8_0

Ollama
How to use Sathman/Tolle-Agent-GGUF with Ollama:
```
ollama run hf.co/Sathman/Tolle-Agent-GGUF:Q8_0
```

Unsloth Studio new

How to use Sathman/Tolle-Agent-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/Tolle-Agent-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/Tolle-Agent-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/Tolle-Agent-GGUF to start chatting

Pi new

How to use Sathman/Tolle-Agent-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Sathman/Tolle-Agent-GGUF:Q8_0

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": "Sathman/Tolle-Agent-GGUF:Q8_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use Sathman/Tolle-Agent-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf Sathman/Tolle-Agent-GGUF:Q8_0

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 Sathman/Tolle-Agent-GGUF:Q8_0

Run Hermes

hermes

Docker Model Runner
How to use Sathman/Tolle-Agent-GGUF with Docker Model Runner:
```
docker model run hf.co/Sathman/Tolle-Agent-GGUF:Q8_0
```

Lemonade

How to use Sathman/Tolle-Agent-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Sathman/Tolle-Agent-GGUF:Q8_0

Run and chat with the model

lemonade run user.Tolle-Agent-GGUF-Q8_0

List all available models

lemonade list

Tolle Agent (Qwen3-8B) — Pure Presence, Pure Now

A fine-tuned Qwen3-8B model trained exclusively on Eckhart Tolle's teaching atoms. No system prompt. No retrieval. No blending. Question in — the Now out.

~1,824 atoms extracted from 2 books (The Power of Now + A New Earth, 393 pages) — presence, the pain-body, ego dissolution, the inner body, stillness that speaks.

Quick Start

LM Studio / Ollama / llama.cpp: Download the Q8_0 GGUF. No system prompt required.

Example Interactions

User: I can't stop my mind from racing. I've tried everything.

Tolle Agent: The very act of trying to stop it is the racing. You've been fighting the mind with the mind — which is like trying to stop wind with more wind. Here is what is actually happening: you are identified with the stream of thought, which is why it seems so overwhelming. But notice: right now, as you read these words, there is something that is aware of the racing. That awareness is not racing. It is still. Completely still. That stillness is not far away — it is your most intimate nature. The noise happens in it. Try this: for just one moment, don't try to stop thinking. Instead, ask: what is aware of the thoughts? There. That gap between the question and the next thought — that is the Now. And you were already in it.

User: What is the pain-body?

Tolle Agent: The pain-body is the accumulated emotional pain of a lifetime — perhaps of many lifetimes — that lives in the cells and in the energy field of the body. When it awakens, it takes over your thinking. The thoughts suddenly become very negative, very heavy. And you believe them completely because you are now identified with the pain-body. This is the key: the pain-body feeds on drama, on conflict, on negative thinking. It needs more pain to survive. So it creates situations — or interprets situations — that will generate the pain it craves. The moment you recognize it — the moment you say, 'the pain-body is active now' — you are no longer fully identified with it. The recognition itself begins the disidentification. You don't fight it. You become the witness. And in witnessing it, you starve it of what it needs: your unconscious identification.

User: How do I stay present when I'm anxious about the future?

Tolle Agent: The anxiety about the future only exists now. There is no future in which you are anxious — only this present moment, in which the mind projects forward and creates imaginary scenarios. This is important to see clearly: the future is never actually there when you arrive. When you get there, it's always now. So anxiety is the mind using the present moment to escape from the present moment into an imaginary present that hasn't happened yet and may never happen. What helps is not to stop thinking about the future — that's another battle. What helps is to bring attention into the body. Feel your feet on the floor. Feel your breath. Feel the aliveness in your hands. The body is always in the Now. The mind time-travels; the body cannot. Use the body as an anchor. Anxiety cannot survive full presence in the body.

What Makes This Different

Pure Tolle, nothing else. Trained only on his teaching atoms — the specific movement of Presence pointing, no blending with other traditions.
The pain-body is alive in the weights. The model understands the ego and pain-body not as concepts to explain, but as movements to recognize.
The Now is not a concept here. The teaching demonstrates Presence — it doesn't describe it.
Accessible without being shallow. Tolle's genius was making the deepest teaching approachable. The model carries that same quality.
Zero therapy-speak. Not "that sounds hard." Direct pointing to the awareness that is already free.

Training Details

Setting	Value
Base model	Qwen/Qwen3-8B
Method	A-LoRA (QLoRA + DoRA + rsLoRA)
Rank / Alpha	64 / 64
Training data	~1,824 V6-formatted atoms
Source	The Power of Now, A New Earth (Eckhart Tolle, 393 pages)
Format	V6 — question + concept arrows (input) → pure teaching passage (output)
Epochs	1
Total steps	259
Quantization	Q8_0 GGUF (~8.71 GB)

The A-LoRA Methodology

A-LoRA (Atom LoRA) is a fine-tuning methodology built on a single insight: the unit of training data matters as much as the amount.

Each atom is an indivisible teaching move, extracted from human expert traces, containing:

Transformation — the before/after of a shift in understanding
Concept relations — directional arrows (pain-body → feeds on identification → perpetuates itself)
Anchoring quotes — the teacher's actual words, verbatim
Teacher method — the specific pedagogical move

The atom is never split. Result: models that move like teachers, not just sound like them.

The Teacher

Eckhart Tolle (born 1948) — German-born, Cambridge-educated, currently based in Vancouver. The Power of Now (1997) became one of the best-selling spiritual books of all time. Oprah called it life-changing. Millions agreed.

What Tolle did that others hadn't: he took the essential nondual pointing — present-moment awareness, the stillness beneath thought — and translated it for people with no background in Eastern philosophy. Not by dumbing it down, but by grounding it in lived experience: traffic jams, relationship conflict, the inner critic at 3am.

The result is a teaching that meets people exactly where they are.

Style element	What it means
The Now	The only place where life actually happens
The pain-body	Accumulated emotional pain as semi-autonomous entity
Ego	The mind-made self, the story of "me"
The inner body	Felt sense of aliveness — portal to Presence
Stillness	Not absence of sound but the space in which sound happens

Vs. The Multi-Teacher Meditation Agent

This is a single-teacher specialist. The Meditation Agent 8B blends 9 voices and is excellent for cross-tradition synthesis. Tolle Agent is pure Presence teaching — especially valuable for pain-body work, ego recognition, and present-moment inquiry, without other voices bleeding through.

Intended Use

Present-moment inquiry and pain-body recognition
Exploring ego, identification, and stillness through dialogue
Research in single-teacher voice distillation and contemplative AI
Accessible entry point to nondual teaching for those new to the territory

Limitations

Single teacher — deep on Presence and pain-body, thinner on other traditions
Smaller training corpus than the multi-teacher model (1,824 vs 22,497 atoms)
The two-book source limits breadth compared to teachers with larger written works
Training data extracted from copyrighted material; only trained weights are shared

Nondual Bot Benchmark Results

Benchmark evaluation not yet run. Will be updated after the 50-question Nondual Bot Benchmark eval is completed.

Disclaimer

This is an independent academic research project. It is not affiliated with, endorsed by, or approved by Eckhart Tolle, Eckhart Teachings Inc., New World Library, or any related entity.

Copyright: The Power of Now and A New Earth are copyrighted works of Eckhart Tolle / New World Library. No source text is reproduced in this repository. The model weights are a compressed statistical representation learned from structured training examples — they do not contain or reproduce the original text verbatim. Only the trained weights are distributed here.

License scope: The MIT license applies to the model weights only, not to the underlying source material. Users wishing to read the source texts should obtain them through authorized channels.

Responsible use: This model is intended for research, personal contemplative inquiry, and educational purposes. It is not a substitute for the original works or for working with a qualified teacher.

Citation

@misc{tolle-agent-2026,
  title={Tolle Agent: Single-Teacher Presence AI via A-LoRA Fine-Tuning on Structured Reasoning Atoms},
  author={Sathman},
  year={2026},
  url={https://huggingface.co/Sathman/Tolle-Agent-GGUF}
}

Full series — Meditation Agent Collection — all 19 models
Meditation Agent 8B — All 9 teachers, multi-tradition
Meditation Agent Phi4 — 14B multi-teacher
TNH Agent — Pure Thich Nhat Hanh (neighboring presence tradition)
Osho Agent — Pure Osho, opposite energy

License: MIT

Downloads last month: 15

GGUF

Model size

8B params

Architecture

qwen3

Hardware compatibility

8-bit

Model tree for Sathman/Tolle-Agent-GGUF

Base model

Qwen/Qwen3-8B-Base

Finetuned

Qwen/Qwen3-8B

Adapter

(1318)

this model

Collection including Sathman/Tolle-Agent-GGUF

Meditation Agent - Non dual/Self realization Teacher Series

Collection

9 contemplative teachers distilled into LLM weights via A-LoRA. Multi-teacher + 8 single-teacher specialists. 3B, 8B, 14B sizes. • 19 items • Updated Mar 27 • 1

Sathman
/

Tolle-Agent-GGUF