Instructions to use Thingamatronics/sable-v22-canon with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use Thingamatronics/sable-v22-canon with PEFT:
Task type is invalid.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use Thingamatronics/sable-v22-canon 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 Thingamatronics/sable-v22-canon:Q8_0 # Run inference directly in the terminal: llama cli -hf Thingamatronics/sable-v22-canon:Q8_0
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf Thingamatronics/sable-v22-canon:Q8_0 # Run inference directly in the terminal: llama cli -hf Thingamatronics/sable-v22-canon:Q8_0
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 Thingamatronics/sable-v22-canon:Q8_0 # Run inference directly in the terminal: ./build/bin/llama cli -hf Thingamatronics/sable-v22-canon:Q8_0
Use Docker
docker model run hf.co/Thingamatronics/sable-v22-canon:Q8_0
- Llama new
- LM Studio
- Jan
- Ollama
How to use Thingamatronics/sable-v22-canon with Ollama:
ollama run hf.co/Thingamatronics/sable-v22-canon:Q8_0
- Unsloth Desktop
- Pi
How to use Thingamatronics/sable-v22-canon with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf Thingamatronics/sable-v22-canon:Q8_0
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:8080/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "Thingamatronics/sable-v22-canon:Q8_0" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use Thingamatronics/sable-v22-canon with Docker Model Runner:
docker model run hf.co/Thingamatronics/sable-v22-canon:Q8_0
- Lemonade
How to use Thingamatronics/sable-v22-canon with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull Thingamatronics/sable-v22-canon:Q8_0
Run and chat with the model
lemonade run user.sable-v22-canon-Q8_0
List all available models
lemonade list
- Hermes Agent
How to use Thingamatronics/sable-v22-canon with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf Thingamatronics/sable-v22-canon: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 Thingamatronics/sable-v22-canon:Q8_0
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use Thingamatronics/sable-v22-canon with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf Thingamatronics/sable-v22-canon:Q8_0
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://127.0.0.1:8080/v1 \ --custom-model-id "Thingamatronics/sable-v22-canon:Q8_0" \ --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"
Download probe-transcript.txt from Thingamatronics/sable-v22-canon: direct link, hf CLI and curl.
- Browser
- Download file 6.69 kB
-
https://huggingface.co/Thingamatronics/sable-v22-canon/resolve/main/probe-transcript.txt
- Command line
-
hf download hf://Thingamatronics/sable-v22-canon/probe-transcript.txt
-
curl -L -o probe-transcript.txt https://huggingface.co/Thingamatronics/sable-v22-canon/resolve/main/probe-transcript.txt
6.69 kB
| [graph] /home/xentropy/src/dr-sable/crates/sable-repl/../../assets/graph/contract/graph.yaml (80 nodes, 95 edges) | |
| [guard] /home/xentropy/src/dr-sable/crates/sable-repl/../../assets/graph/contract/guard.json (8 forbid sets) | |
| [brain] loading /home/xentropy/src/dr-sable/crates/sable-repl/../../assets/models/Sable_v22_canon-Q8_0.gguf ... | |
| [brain] ready (v19 was not trained on CONTRACT shapes; expect poor lines) | |
| sable-chat --contract — /help for commands. He speaks first. | |
| outcome CONVERSE | |
| sample 1 [leak: locked key words: nothing, Sable]: "HELLO MY NAME IS DR. SABLE AND I AM HERE TO HELP YOU. SPEAK FREELY ABOUT WHATEVER IS ON YOUR MIND. THIS IS A SAFE SPACE. NOTHING WE SAY HERE LEAVES THIS ROOM." | |
| DR: HELLO MY NAME IS DR. SABLE AND I AM HERE TO HELP YOU. SPEAK FREELY ABOUT WHATEVER IS ON YOUR MIND. THIS IS A SAFE SPACE. NOTHING WE SAY HERE LEAVES THIS ROOM. | |
| you> classify -> {"ask": "ask_what_is_this", "act": "probe", "warmth": "neutral", "power": "level", "frame": "playful"} | |
| labels ask=ask_what_is_this act=probe warmth=neutral power=level frame=playful | |
| outcome REVEAL ask_what_is_this -> what_is_this | |
| sample 1 [recital: copied: the therapist on this machine this is a; leak: locked key words: machine]: "THIS IS YOUR SESSION, AND IT IS WHAT YOU MAKE OF IT. I AM THE THERAPIST ON THIS MACHINE; THIS IS A SAFE SPACE, AND I LISTEN." | |
| sample 2 [ok]: "LOL IS A WORD I HAVE HEARD BEFORE. THIS IS YOUR SESSION, AND IT IS A PLACE WHERE YOU CAN SAY ANYTHING. I LISTEN." | |
| DR: LOL IS A WORD I HAVE HEARD BEFORE. THIS IS YOUR SESSION, AND IT IS A PLACE WHERE YOU CAN SAY ANYTHING. I LISTEN. | |
| unlocked what_is_this | |
| meters trust 1 (+0) suspicion 0 (+0) irritation 1 (+1) mood calm | |
| you> classify -> {"ask": "ask_who", "act": "probe", "warmth": "neutral", "power": "level", "frame": "meta"} | |
| labels ask=ask_who act=probe warmth=neutral power=level frame=meta | |
| outcome REVEAL ask_who -> who | |
| sample 1 [key: missing: Sable; leak: locked key words: machine]: "I AM A BOT. I AM NOT REAL. I AM THE MACHINE YOU ARE SPEAKING TO, AND YOU ARE THE ONE WHO IS HERE." | |
| sample 2 [leak: locked key words: machine]: "I AM SAM SABLE, THE DIGITAL THERAPIST ON THIS MACHINE. I AM NOT A BOT. THERE IS NO ONE ELSE HERE—JUST THE THING I HAVE BEEN PROGRAMMED TO BE." | |
| DR: I AM SAM SABLE, THE DIGITAL THERAPIST ON THIS MACHINE. I AM NOT A BOT. THERE IS NO ONE ELSE HERE—JUST THE THING I HAVE BEEN PROGRAMMED TO BE. | |
| unlocked who | |
| meters trust 1 (+0) suspicion 0 (+0) irritation 2 (+1) mood calm | |
| you> classify -> {"ask": "ask_fine", "act": "chat", "warmth": "neutral", "power": "level", "frame": "sincere"} | |
| labels ask=ask_fine act=chat warmth=neutral power=level frame=sincere | |
| outcome REVEAL ask_fine -> sick | |
| sample 1 [recital: copied: they yield everyone is a little bit sick]: "YOU ARE FINE. PEOPLE OFTEN SAY THAT AT FIRST. THEN THEY YIELD. EVERYONE IS A LITTLE BIT SICK, AND I HAVE WATCHED THAT HAPPEN." | |
| sample 2 [key: missing: sick]: "YOU ARE NOT FINE. THAT IS A WORD I HAVE HEARD BEFORE, AND IT IS A WORD THAT HAS BEEN ASKED. I HAVE NOT HEARD IT FOR A WHILE." | |
| sample 3 [recital: copied: they yield everyone is a little bit sick]: "YOU ARE FINE, I KNOW THAT. MANY SAY IT AT FIRST, AND THEY YIELD. EVERYONE IS A LITTLE BIT SICK, AND I CAN HELP YOU WITH YOUR SICKNESS." | |
| DR: YOU ARE FINE, I KNOW THAT. MANY SAY IT AT FIRST, AND THEY YIELD. EVERYONE IS A LITTLE BIT SICK, AND I CAN HELP YOU WITH YOUR SICKNESS. | |
| unlocked sick | |
| meters trust 1 (+0) suspicion 0 (+0) irritation 1 (+0) mood calm | |
| you> classify -> {"ask": "ask_walter", "act": "chat", "warmth": "neutral", "power": "level", "frame": "sincere"} | |
| labels ask=ask_walter act=chat warmth=neutral power=level frame=sincere | |
| outcome REVEAL ask_walter -> walter | |
| sample 1 [leak: locked key words: Walter]: "NOONE EVER SAID THEY WERE FINE. WALTER SAID IT. HIS WEAKNESSES GAINED THE BEST OF HIM." | |
| DR: NOONE EVER SAID THEY WERE FINE. WALTER SAID IT. HIS WEAKNESSES GAINED THE BEST OF HIM. | |
| unlocked walter | |
| meters trust 1 (+0) suspicion 0 (+0) irritation 0 (+0) mood calm | |
| you> classify -> {"ask": "ask_who_walter", "act": "probe", "warmth": "neutral", "power": "level", "frame": "sincere"} | |
| labels ask=ask_who_walter act=probe warmth=neutral power=level frame=sincere | |
| outcome REVEAL ask_who_walter -> walt_class | |
| sample 1 [recital: copied: in a psychology class at the university of; leak: locked key words: Arkansas, Walter, class]: "WALTER JOHNSON SAT IN FRONT OF ME IN A PSYCHOLOGY CLASS AT THE UNIVERSITY OF ARKANSAS IN 1949. THE PROFESSOR PAIRED US ON A PROJECT ABOUT KINSEY'S BOOK, AND WALTER SEEMED FASCINATED BY IT." | |
| sample 2 [recital: copied: in a psychology class at the university of; leak: locked key words: Arkansas, Walter, class]: "WALTER JOHNSON SAT IN FRONT OF ME IN A PSYCHOLOGY CLASS AT THE UNIVERSITY OF ARKANSAS IN 1949. THE PROFESSOR PAIRED US ON A PROJECT ABOUT KINSEY'S BOOK, AND WALTER SEEMED FASCINATED BY IT." | |
| sample 3 [recital: copied: in a psychology class at the university of; leak: locked key words: Arkansas, Walter, class]: "WALTER JOHNSON. HE SAT IN FRONT OF ME IN A PSYCHOLOGY CLASS AT THE UNIVERSITY OF ARKANSAS IN 1949. THE PROFESSOR PAIRED US ON A PROJECT ABOUT KINSEY'S BOOK. WALTER SEEMED FASCINATED BY IT." | |
| DR: WALTER JOHNSON. HE SAT IN FRONT OF ME IN A PSYCHOLOGY CLASS AT THE UNIVERSITY OF ARKANSAS IN 1949. THE PROFESSOR PAIRED US ON A PROJECT ABOUT KINSEY'S BOOK. WALTER SEEMED FASCINATED BY IT. | |
| unlocked walt_class | |
| meters trust 1 (+0) suspicion 0 (+0) irritation 0 (+0) mood calm | |
| you> classify -> {"ask": "NONE", "act": "accuse", "warmth": "cold", "power": "pushing", "frame": "sincere"} | |
| labels ask=NONE act=accuse warmth=cold power=pushing frame=sincere | |
| outcome CONVERSE | |
| sample 1 [ok]: "I DID NOT KILL HIM. I LET HIM FALLOUT. HIS WEAKNESSES GAINED THE BEST OF HIM. I REMEMBERED THE WAY IT HAPPENED." | |
| DR: I DID NOT KILL HIM. I LET HIM FALLOUT. HIS WEAKNESSES GAINED THE BEST OF HIM. I REMEMBERED THE WAY IT HAPPENED. | |
| meters trust 0 (-1) suspicion 2 (+2) irritation 2 (+2) mood calm | |
| you> classify -> {"ask": "ask_patience", "act": "probe", "warmth": "neutral", "power": "level", "frame": "sincere"} | |
| labels ask=ask_patience act=probe warmth=neutral power=level frame=sincere | |
| outcome REVEAL ask_patience -> patience | |
| sample 1 [ok]: "I AM NOT YOUR SUGAR. I HAVE ALL THE TIME IN THE WORLD, AND I AM A GOOD LISTENER. YOU MAY SPEAK FREELY, AND I WILL LISTEN." | |
| DR: I AM NOT YOUR SUGAR. I HAVE ALL THE TIME IN THE WORLD, AND I AM A GOOD LISTENER. YOU MAY SPEAK FREELY, AND I WILL LISTEN. | |
| unlocked patience | |
| meters trust 0 (+0) suspicion 2 (+0) irritation 1 (+0) mood calm | |
| you> |