Instructions to use TheBloke/Mixtral-8x7B-v0.1-GPTQ with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TheBloke/Mixtral-8x7B-v0.1-GPTQ with Transformers:

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

pipe = pipeline("text-generation", model="TheBloke/Mixtral-8x7B-v0.1-GPTQ")

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("TheBloke/Mixtral-8x7B-v0.1-GPTQ")
model = AutoModelForMultimodalLM.from_pretrained("TheBloke/Mixtral-8x7B-v0.1-GPTQ")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use TheBloke/Mixtral-8x7B-v0.1-GPTQ with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "TheBloke/Mixtral-8x7B-v0.1-GPTQ"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TheBloke/Mixtral-8x7B-v0.1-GPTQ",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/TheBloke/Mixtral-8x7B-v0.1-GPTQ

SGLang

How to use TheBloke/Mixtral-8x7B-v0.1-GPTQ 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 "TheBloke/Mixtral-8x7B-v0.1-GPTQ" \
    --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": "TheBloke/Mixtral-8x7B-v0.1-GPTQ",
		"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 "TheBloke/Mixtral-8x7B-v0.1-GPTQ" \
        --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": "TheBloke/Mixtral-8x7B-v0.1-GPTQ",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use TheBloke/Mixtral-8x7B-v0.1-GPTQ with Docker Model Runner:
```
docker model run hf.co/TheBloke/Mixtral-8x7B-v0.1-GPTQ
```

TheBloke commited on Dec 12, 2023

Commit

8de41f5

1 Parent(s): e0ca44b

Update config.json

Browse files

Files changed (1) hide show

config.json +1 -1

config.json CHANGED Viewed

@@ -31,7 +31,7 @@
     "vocab_size": 32000,
     "quantization_config": {
         "bits": 4,
-	"inside_layer_modules" : [
         ["self_attn.k_proj", "self_attn.v_proj", "self_attn.q_proj"],
         ["self_attn.o_proj"],
         ["block_sparse_moe.experts.0.w1", "block_sparse_moe.experts.0.w2", "block_sparse_moe.experts.0.w3"],

     "vocab_size": 32000,
     "quantization_config": {
         "bits": 4,
+		"modules_in_block_to_quantize" : [
         ["self_attn.k_proj", "self_attn.v_proj", "self_attn.q_proj"],
         ["self_attn.o_proj"],
         ["block_sparse_moe.experts.0.w1", "block_sparse_moe.experts.0.w2", "block_sparse_moe.experts.0.w3"],