Instructions to use MaziyarPanahi/WizardLM-2-8x22B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MaziyarPanahi/WizardLM-2-8x22B-GGUF with Transformers:

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

pipe = pipeline("text-generation", model="MaziyarPanahi/WizardLM-2-8x22B-GGUF")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("MaziyarPanahi/WizardLM-2-8x22B-GGUF", dtype="auto")

llama-cpp-python

How to use MaziyarPanahi/WizardLM-2-8x22B-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="MaziyarPanahi/WizardLM-2-8x22B-GGUF",
	filename="WizardLM-2-8x22B.IQ1_M.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use MaziyarPanahi/WizardLM-2-8x22B-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf MaziyarPanahi/WizardLM-2-8x22B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf MaziyarPanahi/WizardLM-2-8x22B-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf MaziyarPanahi/WizardLM-2-8x22B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf MaziyarPanahi/WizardLM-2-8x22B-GGUF: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 MaziyarPanahi/WizardLM-2-8x22B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf MaziyarPanahi/WizardLM-2-8x22B-GGUF: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 MaziyarPanahi/WizardLM-2-8x22B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf MaziyarPanahi/WizardLM-2-8x22B-GGUF:Q4_K_M

Use Docker

docker model run hf.co/MaziyarPanahi/WizardLM-2-8x22B-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use MaziyarPanahi/WizardLM-2-8x22B-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MaziyarPanahi/WizardLM-2-8x22B-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MaziyarPanahi/WizardLM-2-8x22B-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/MaziyarPanahi/WizardLM-2-8x22B-GGUF:Q4_K_M

SGLang

How to use MaziyarPanahi/WizardLM-2-8x22B-GGUF 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 "MaziyarPanahi/WizardLM-2-8x22B-GGUF" \
    --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": "MaziyarPanahi/WizardLM-2-8x22B-GGUF",
		"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 "MaziyarPanahi/WizardLM-2-8x22B-GGUF" \
        --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": "MaziyarPanahi/WizardLM-2-8x22B-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Ollama
How to use MaziyarPanahi/WizardLM-2-8x22B-GGUF with Ollama:
```
ollama run hf.co/MaziyarPanahi/WizardLM-2-8x22B-GGUF:Q4_K_M
```

Unsloth Studio

How to use MaziyarPanahi/WizardLM-2-8x22B-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for MaziyarPanahi/WizardLM-2-8x22B-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for MaziyarPanahi/WizardLM-2-8x22B-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for MaziyarPanahi/WizardLM-2-8x22B-GGUF to start chatting

Docker Model Runner
How to use MaziyarPanahi/WizardLM-2-8x22B-GGUF with Docker Model Runner:
```
docker model run hf.co/MaziyarPanahi/WizardLM-2-8x22B-GGUF:Q4_K_M
```

Lemonade

How to use MaziyarPanahi/WizardLM-2-8x22B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull MaziyarPanahi/WizardLM-2-8x22B-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.WizardLM-2-8x22B-GGUF-Q4_K_M

List all available models

lemonade list

Help regarding best quantization for below PC specification.

#15

by bikkikumarsha - opened Apr 18, 2024

Discussion

bikkikumarsha

Apr 18, 2024

Since I don't want to spend hours download something that doesn't run. I am wondering which quantization would fit nicely in my system.
CPU: Intel core i7 13700k
Ram: 64GB
GPU: RTX 3090 24GB dedicated memory

Also is there any general rule of thumb that we should be following.

MaziyarPanahi

Owner Apr 18, 2024

Hi,
I am about to upload the IQ-1 models, which are the smallest models. Those should fit without any issue.

bikkikumarsha

Apr 18, 2024

Thanks for all the good work..

tsalvoch

Apr 18, 2024

Hello,

I have roughly the same setup as you (RTX3090, 64GB RAM, Intel Core i9) and for my testing, I used TextGenerationWebui and the IQ1_M model loaded with 36 layers on the GPU.

Also, I had to limit the context size to 4096 and it seems that the max_new_tokens value has an impact on the quality of the results. I get better results with a max_new_tokens of 512 than with 1024.

However, it is very slow, I only get 1.6 tokens/s, so not really usable due to the delays.

I did a quick test last night with an RTX6000 which allows this model to be fully loaded in VRAM and I was getting around 25 tokens/s. In my opinion, this model requires more power than what a standard gaming PC has.

MaziyarPanahi

Owner Apr 18, 2024

Thanks @tsalvoch for sharing your setup, that's a big help to others.

Adzeiros

Aug 8, 2024

@MaziyarPanahi Just curious, is there a quant worth running on dual 3090's? Preferably fully loaded into VRAM with 8-32k context (highest possible)... or would you need at least 3 3090's to run a quant that's worth it?

MaziyarPanahi

Owner Aug 8, 2024

@Adzeiros depends on the model to be honest. if we are talking about this specific model, it's pretty huge! so even a Q3 would do a good job and still pretty beefy. I've seen people using 2 3090 with some offloading, but they were happy for their use cases. (proofreading, rewriting, etc.)
Based on the size of the model and the tasks you require, it should be possible to tradeoff and find a middle ground for 2 or 3 3090

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment