Instructions to use nkpz/llama2-22b-frankenwizard with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nkpz/llama2-22b-frankenwizard with Transformers:

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

pipe = pipeline("text-generation", model="nkpz/llama2-22b-frankenwizard")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("nkpz/llama2-22b-frankenwizard")
model = AutoModelForCausalLM.from_pretrained("nkpz/llama2-22b-frankenwizard")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use nkpz/llama2-22b-frankenwizard with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "nkpz/llama2-22b-frankenwizard"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nkpz/llama2-22b-frankenwizard",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/nkpz/llama2-22b-frankenwizard

SGLang

How to use nkpz/llama2-22b-frankenwizard 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 "nkpz/llama2-22b-frankenwizard" \
    --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": "nkpz/llama2-22b-frankenwizard",
		"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 "nkpz/llama2-22b-frankenwizard" \
        --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": "nkpz/llama2-22b-frankenwizard",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use nkpz/llama2-22b-frankenwizard with Docker Model Runner:
```
docker model run hf.co/nkpz/llama2-22b-frankenwizard
```

Thanks to chargoddard for the original 22b model and merge script: https://huggingface.co/chargoddard/llama2-22b

This uses https://huggingface.co/Blackroot/FrankensteinsMonster-13B as a base, with https://huggingface.co/ehartford/WizardLM-33B-V1.0-Uncensored as the donor model.

This an experimental model and it sometimes forgets its context or says something nonsensical, but maybe some further fine tuning would smooth things out. This has less guardrails than my prior merge, https://huggingface.co/nkpz/llama2-22b-chat-wizard-uncensored

Took around 2 hours to merge with 32gb of ram and about 115gb of swap used.

Downloads last month: 4

Safetensors

Model size

22B params

Tensor type

F32

F16