Instructions to use Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-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 Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-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 Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF:Q5_K_M # Run inference directly in the terminal: llama cli -hf Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF:Q5_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF:Q5_K_M # Run inference directly in the terminal: llama cli -hf Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-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/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF:Q5_K_M # Run inference directly in the terminal: ./llama-cli -hf Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-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/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF:Q5_K_M # Run inference directly in the terminal: ./build/bin/llama-cli -hf Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF:Q5_K_M
Use Docker
docker model run hf.co/Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF:Q5_K_M
- LM Studio
- Jan
- vLLM
How to use Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-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/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF:Q5_K_M
- Ollama
How to use Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF with Ollama:
ollama run hf.co/Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF:Q5_K_M
- Unsloth Desktop
- Docker Model Runner
How to use Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF with Docker Model Runner:
docker model run hf.co/Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF:Q5_K_M
- Lemonade
How to use Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF:Q5_K_M
Run and chat with the model
lemonade run user.Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF-Q5_K_M
List all available models
lemonade list
- Atomic Chat
Install from WinGet (Windows)
winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF:Q5_K_M# Run inference directly in the terminal:
llama cli -hf Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF:Q5_K_MUse 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/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF:Q5_K_M# Run inference directly in the terminal:
./llama-cli -hf Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF:Q5_K_MBuild 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/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF:Q5_K_M# Run inference directly in the terminal:
./build/bin/llama-cli -hf Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF:Q5_K_MUse Docker
docker model run hf.co/Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF:Q5_K_MMeditation Agent: Atmananda v2 (SmolLM3 3B)
A lightweight single-teacher contemplative model from the Meditation Agent v2 line for Atmananda.
This is the v2 successor to Atmananda-Agent-SmolLM3-3B-GGUF. It is trained on the newer E-atoms corpus, where the target is teaching_e, not the older atom surface.
Collections:
- New v2 collection: Meditation Agent v2 - EAtoms Teacher Series
- Previous collection: Meditation Agent - Non-Dual Self-Realization Teacher Series
Why Download This
- Single-teacher branch. This model is tuned specifically for Atmananda, not the mixed multi-teacher voice.
- Small local footprint. The current release file is
Q5_K_M, which runs comfortably on local setups that cannot host larger models. - Newer v2 corpus. The
E-atomsline is sharper, less synthetic, and less likely to collapse into generic spiritual assistant prose.
Teacher Signal
This branch is tuned for lucid direct-path discrimination, subtle reasoning, and immediate recognition of awareness.
Sample Outputs
These are from the local 5q smoke eval on the finished merged model.
Example 1
Prompt: How does fear relate to thought?
Fear and thought are like two rivers flowing from the same source, yet they follow different courses. You feel fear--that gripping sensation in your chest, that surge of panic--and then you notice it, become aware of it; this noticing itself is thought. The fear comes first as an immediate response to danger or uncertainty, but awareness follows close behind, dividing what was once undivided: the raw emotion and its recognition. Thought is simply memory at work here--it takes hold when you try to grasp what's happening instead of allowing yourself to fully experience both without naming them. Fear and thought are not separate things fighting for dominance within you; rather, they are aspects of consciousness itself, each serving a necessary function while remaining bound by their relationship with awareness.
Example 2
Prompt: Is there a self?
Look closely at what you're asking. The question itself assumes something to be found, some hidden truth waiting beneath the surface of experience--a private'self' that might be discovered or confirmed through introspection.
But consider this: when you say 'is there a self?', who is it that makes such a statement? Is there an observer separate from observation? A thinker distinct from thought?
The seeking itself reveals where your attention actually rests. You cannot find what you are because finding requires a finder. And if no finder exists, how could anything be found?
Example 3
Prompt: What do I do with anger when it arises?
When anger surfaces--that burning sensation in the chest, that tightening of muscles around the throat--you have two choices: you can either surrender to its force or meet it directly. Anger is like water flowing through a dam; it will find a way out whether you resist or not. But if you face it openly without flinching, something remarkable happens. The heat doesn't disappear entirely, but it shifts. The fire loses its sharpness and becomes more diffuse. You're still angry, yes? But now there's also an awareness watching over what's happening within you. This isn't suppression--it's acceptance without identification. When you see your anger clearly rather than running from it, something essential transforms. The energy that once seemed so dangerous begins to dissolve into the very ground from which it came. Not by fighting against itself, but simply by recognizing it for what it truly is: movement within consciousness itself.
Training Summary
- Base model:
HuggingFaceTB/SmolLM3-3B-Base - Training format:
V6E - Teacher examples:
2,798 - Adapter recipe:
QDoRA + rsLoRA - Rank:
32 - Epochs:
1 - Completion-only loss: enabled
Run Metrics
- Run name:
teacher-smollm3-v6e-atmananda-l2-r32 - Training runtime:
24.6 minutes - Final train loss:
1.7902 - Final eval loss:
1.7965 - Final eval token accuracy:
0.5507
Files
Atmananda_Agent-SmolLM3-3B-v2-Q5_K_M.gguf(2.2 GB)
Honest Limits
- This is still a compact
3Bmodel, so some broader contemplative phrasing can leak into non-teacher-specific prompts. - The
5qsmoke eval is encouraging, but it is not a full benchmark or final quality proof.
Public Naming
- Public repo:
Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF - Internal model folder:
Atmananda_Agent-SmolLM3-3B-v2
- Downloads last month
- 17
5-bit
Model tree for Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF
Base model
HuggingFaceTB/SmolLM3-3B-Base
Install (macOS, Linux)
# Start a local OpenAI-compatible server with a web UI: llama serve -hf Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF:Q5_K_M# Run inference directly in the terminal: llama cli -hf Sathman/Meditation-Agent-Atmananda-SmolLM3-3B-v2-GGUF:Q5_K_M