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

Libraries

How to use KaraKaraWarehouse/PygKiCOTlion with Transformers:

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

pipe = pipeline("text-generation", model="KaraKaraWarehouse/PygKiCOTlion")

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("KaraKaraWarehouse/PygKiCOTlion")
model = AutoModelForMultimodalLM.from_pretrained("KaraKaraWarehouse/PygKiCOTlion")

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use KaraKaraWarehouse/PygKiCOTlion with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/KaraKaraWarehouse/PygKiCOTlion

SGLang

How to use KaraKaraWarehouse/PygKiCOTlion 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 "KaraKaraWarehouse/PygKiCOTlion" \
    --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": "KaraKaraWarehouse/PygKiCOTlion",
		"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 "KaraKaraWarehouse/PygKiCOTlion" \
        --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": "KaraKaraWarehouse/PygKiCOTlion",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use KaraKaraWarehouse/PygKiCOTlion with Docker Model Runner:
```
docker model run hf.co/KaraKaraWarehouse/PygKiCOTlion
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Model Card for PygKiCOTlion

PygKiCOTlion is a a lora merge of Pygmalion-2-13b-SuperCOT + Kimiko v2

Model Details

Q: "Why do you do this?!"
A: Was bored.

Model Description

Developed by: KaraKaraWitch (Merge), kaiokendev (Original SuperCOT LoRA), nRuaif (Kimiko v2 LoRA), kingbri (Pygmalion 2 13b SuperCOT)
Model type: Decoder only
License: LLaMA2 (PygKiCOTlion), SuperCOT (MIT), Kimiko v2 (CC BY-NC-SA (?))
Finetuned from model [optional]: LLaMA2

Model Sources [optional]

Uses

YYMV.

Direct Use

Usage:

Since this is a merge between Pygmalion 2 13b SuperCOT and Kimiko v2, the following instruction formats should work:

Metharme:

<|system|>Your system prompt goes here.<|user|>Are you alive?<|model|>

Alpaca:

### Instruction:
Your instruction or question here.
### Response:

Bias, Risks, and Limitations

YMMV. This is untested territory.

Testing Feedbakc

Notes from KaraKaraWitch:

The model feels weirdly loopy compared to MythKiCOTlion at lower temps.
Higher temps the model tries to venture out of it's comfort zone at the cost of making it not stick to the model card as close as expected.

Training Details

N/A. Refer to the respective LoRa's and models.

Downloads last month: 3

Model tree for KaraKaraWarehouse/PygKiCOTlion

Quantizations

2 models