Instructions to use VAGOsolutions/SauerkrautLM-Mixtral-8x7B-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use VAGOsolutions/SauerkrautLM-Mixtral-8x7B-Instruct with Transformers:

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

pipe = pipeline("text-generation", model="VAGOsolutions/SauerkrautLM-Mixtral-8x7B-Instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("VAGOsolutions/SauerkrautLM-Mixtral-8x7B-Instruct")
model = AutoModelForCausalLM.from_pretrained("VAGOsolutions/SauerkrautLM-Mixtral-8x7B-Instruct", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use VAGOsolutions/SauerkrautLM-Mixtral-8x7B-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "VAGOsolutions/SauerkrautLM-Mixtral-8x7B-Instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "VAGOsolutions/SauerkrautLM-Mixtral-8x7B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/VAGOsolutions/SauerkrautLM-Mixtral-8x7B-Instruct

SGLang

How to use VAGOsolutions/SauerkrautLM-Mixtral-8x7B-Instruct 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 "VAGOsolutions/SauerkrautLM-Mixtral-8x7B-Instruct" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "VAGOsolutions/SauerkrautLM-Mixtral-8x7B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "VAGOsolutions/SauerkrautLM-Mixtral-8x7B-Instruct" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "VAGOsolutions/SauerkrautLM-Mixtral-8x7B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use VAGOsolutions/SauerkrautLM-Mixtral-8x7B-Instruct with Docker Model Runner:
```
docker model run hf.co/VAGOsolutions/SauerkrautLM-Mixtral-8x7B-Instruct
```

Clarification needed and request for GGUF support for llama.cpp?

by rboehme86 - opened Apr 7, 2024

Discussion

rboehme86

Apr 7, 2024

Dear VAGOsolutions

If possible it would be great if you can provide two short responses:

1.) Is you model based on Mixtral 8x7B v0.1 or v0.2? The later was released around the time your model came out and contains some bugfixes. I observed some issues with your model in providing unecessary user example output in some cases and there is the chance that this is related to this.

2.) Could you mabye provide a GGUF version for llama.cpp inference of your model? This should be a quick task for you and would drastically increase your user base.

Thanks a lot - any feedback is welcome,

Robert

DaryoushV

VAGO solutions org Apr 11, 2024

Hi Robert,

The model was fine-tuned on the 0.1 version. Is there really a 0.2v of the Mixtral? I don't think so
You find GGUF Versions for most of our models in our organization card (Thanks to the bloke, who provided these quantizations)

best regards
Daryoush

DavidGF changed discussion status to closed Apr 27, 2024

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