Instructions to use v2ray/Mixtral-8x22B-v0.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use v2ray/Mixtral-8x22B-v0.1 with Transformers:

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

pipe = pipeline("text-generation", model="v2ray/Mixtral-8x22B-v0.1")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("v2ray/Mixtral-8x22B-v0.1")
model = AutoModelForCausalLM.from_pretrained("v2ray/Mixtral-8x22B-v0.1", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use v2ray/Mixtral-8x22B-v0.1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "v2ray/Mixtral-8x22B-v0.1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "v2ray/Mixtral-8x22B-v0.1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/v2ray/Mixtral-8x22B-v0.1

SGLang

How to use v2ray/Mixtral-8x22B-v0.1 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 "v2ray/Mixtral-8x22B-v0.1" \
    --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": "v2ray/Mixtral-8x22B-v0.1",
		"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 "v2ray/Mixtral-8x22B-v0.1" \
        --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": "v2ray/Mixtral-8x22B-v0.1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use v2ray/Mixtral-8x22B-v0.1 with Docker Model Runner:
```
docker model run hf.co/v2ray/Mixtral-8x22B-v0.1
```

Quantized models in GGUF

pinned

by MaziyarPanahi - opened Apr 10, 2024

Discussion

MaziyarPanahi

Apr 10, 2024

Hi @v2ray

Thanks for converting and sharing this model, kudos! I am in the process of uploading GGUF models from 16bit all the way down to 2bit, for those with low resources:

https://huggingface.co/MaziyarPanahi/Mixtral-8x22B-v0.1-GGUF

v2ray

Owner Apr 10, 2024

owo

v2ray changed discussion status to closed Apr 10, 2024

v2ray pinned discussion Apr 10, 2024

v2ray changed discussion status to open Apr 10, 2024

philippemarques

Apr 11, 2024

•

edited Apr 11, 2024

Hello all,

In LMStudio 0.2.18 : "llama.cpp error: 'illegal split file: 4, model must be loaded with the first split'" for the Q4
Any tip to solve this ?
Thank you.

MaziyarPanahi

Apr 11, 2024

Does LM Studio supports this new loading split?

MaziyarPanahi

Apr 13, 2024

Does LM Studio supports this new loading split?

It does support, I can confirm 😊

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