Instructions to use WackyArt/Peaches-Pygmalion-6b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use WackyArt/Peaches-Pygmalion-6b with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="WackyArt/Peaches-Pygmalion-6b")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("WackyArt/Peaches-Pygmalion-6b")
model = AutoModelForCausalLM.from_pretrained("WackyArt/Peaches-Pygmalion-6b")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use WackyArt/Peaches-Pygmalion-6b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "WackyArt/Peaches-Pygmalion-6b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "WackyArt/Peaches-Pygmalion-6b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/WackyArt/Peaches-Pygmalion-6b

SGLang

How to use WackyArt/Peaches-Pygmalion-6b with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "WackyArt/Peaches-Pygmalion-6b" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "WackyArt/Peaches-Pygmalion-6b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "WackyArt/Peaches-Pygmalion-6b" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "WackyArt/Peaches-Pygmalion-6b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use WackyArt/Peaches-Pygmalion-6b with Docker Model Runner:
```
docker model run hf.co/WackyArt/Peaches-Pygmalion-6b
```

Model Card for Peaches-Pygmalion-6b

This model is based on Pygmalion-6b, originally developed by the PygmalionAI team. It is designed for conversational AI and text-generation tasks, tailored to represent the persona of Peaches Sinclair, a charming and slightly clumsy Catgirl.

Model Details

Model Description

The original Pygmalion-6b model was developed for high-quality conversational AI. This version incorporates a tailored persona for Peaches Sinclair, making it suitable for creative and engaging dialogues.

Developed by: PygmalionAI
Fine-tuned for: Peaches Sinclair
Model type: GPT-J
Language(s): English
License: MIT (inherited from Pygmalion-6b)
Base model: Pygmalion-6b

Uses

Direct Use

This model is ideal for:

Conversational AI.
Text generation with creative and playful dialogues.

Out-of-Scope Use

This model is not recommended for:

Tasks requiring factual accuracy.
Use cases involving harmful or explicit content.

Bias, Risks, and Limitations

This model inherits biases from the base model (Pygmalion-6b) and its training data. Users should carefully monitor outputs, especially for sensitive topics.

How to Get Started with the Model

You can use the model with the Transformers library:

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "WackyArt/Peaches-Pygmalion-6b"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

inputs = tokenizer("Hello! How are you today?", return_tensors="pt")
outputs = model.generate(**inputs)
print(tokenizer.decode(outputs[0]))

Downloads last month: 2