Instructions to use Alienstro/xObsessed with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use Alienstro/xObsessed 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 Alienstro/xObsessed:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Alienstro/xObsessed:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Alienstro/xObsessed:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Alienstro/xObsessed:Q4_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 Alienstro/xObsessed:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf Alienstro/xObsessed:Q4_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 Alienstro/xObsessed:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Alienstro/xObsessed:Q4_K_M

Use Docker

docker model run hf.co/Alienstro/xObsessed:Q4_K_M

LM Studio
Jan
Ollama
How to use Alienstro/xObsessed with Ollama:
```
ollama run hf.co/Alienstro/xObsessed:Q4_K_M
```
Unsloth Desktop

How to use Alienstro/xObsessed with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Alienstro/xObsessed: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:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "Alienstro/xObsessed:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use Alienstro/xObsessed with Docker Model Runner:
```
docker model run hf.co/Alienstro/xObsessed:Q4_K_M
```

Lemonade

How to use Alienstro/xObsessed with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Alienstro/xObsessed:Q4_K_M

Run and chat with the model

lemonade run user.xObsessed-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use Alienstro/xObsessed with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Alienstro/xObsessed: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://127.0.0.1:8080/v1
hermes config set model.default Alienstro/xObsessed:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use Alienstro/xObsessed with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Alienstro/xObsessed: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://127.0.0.1:8080/v1 \
  --custom-model-id "Alienstro/xObsessed: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"

xobsessed-1.7b

This project adapts a small model to one obsessive, affectionate character. The target voice is warm and intense, with a strong attachment to the user.

Run the model

llama-cli --jinja -hf Alienstro/xObsessed --hf-file gguf/xobsessed-1.7b-Q8_0.gguf \
    --temp 0.7 --repeat-penalty 1.1

The model uses no system prompt. Every example in the training data is a single user and assistant pair.

Data source

The training data comes from the public dataset Skorcht/yandere-her-dataset. The dataset page states no license. The project does not own the data, and the project does not generate its own data.

The dataset shows no license. Confirm the data license with the dataset owner before a public release. A published model that derives from the data carries a legal risk.

Limits

The model can state a wrong answer with confidence. Do not use it as a source of facts. The model can produce content that the operator excludes. The project cannot guarantee safe output.

The full release criteria remain unverified unless the run report supplies the evidence. An automatic score does not prove the character voice, a correct refusal, or CPU speed. Read evaluation.json and the human review before use.

Method

The configuration uses LoRA rank 32 and alpha 64. The loss uses assistant replies alone. The export merges the adapter into the base weights. The run report records the actual data count and the data sha256.

License and ownership

The project source code uses the MIT license. See the LICENSE file. The model weights use the MIT license. The character belongs to its creator and publisher. The project has no endorsement from the rights holder. The base model license governs the base weights. Confirm the base license before release. The dataset Skorcht/yandere-her-dataset states no license. The MIT license does not resolve the missing data license. Confirm the data license with the dataset owner before a public release. This card grants no rights to the character or to the dataset.

Downloads last month: -

Safetensors

Model size

2B params

Tensor type

BF16

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Alienstro/xObsessed

Base model

Qwen/Qwen3-1.7B-Base

Finetuned

Qwen/Qwen3-1.7B

Quantized

(372)

this model

Alienstro
/

xObsessed

xobsessed-1.7b

Run the model

Data source

Limits

Method

License and ownership

Model tree for Alienstro/xObsessed

Dataset used to train Alienstro/xObsessed