Instructions to use TheBloke/dolphin-2.5-mixtral-8x7b-GPTQ with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TheBloke/dolphin-2.5-mixtral-8x7b-GPTQ with Transformers:

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

pipe = pipeline("text-generation", model="TheBloke/dolphin-2.5-mixtral-8x7b-GPTQ")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("TheBloke/dolphin-2.5-mixtral-8x7b-GPTQ")
model = AutoModelForCausalLM.from_pretrained("TheBloke/dolphin-2.5-mixtral-8x7b-GPTQ")
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

vLLM

How to use TheBloke/dolphin-2.5-mixtral-8x7b-GPTQ with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "TheBloke/dolphin-2.5-mixtral-8x7b-GPTQ"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TheBloke/dolphin-2.5-mixtral-8x7b-GPTQ",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/TheBloke/dolphin-2.5-mixtral-8x7b-GPTQ

SGLang

How to use TheBloke/dolphin-2.5-mixtral-8x7b-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/dolphin-2.5-mixtral-8x7b-GPTQ" \
    --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": "TheBloke/dolphin-2.5-mixtral-8x7b-GPTQ",
		"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 "TheBloke/dolphin-2.5-mixtral-8x7b-GPTQ" \
        --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": "TheBloke/dolphin-2.5-mixtral-8x7b-GPTQ",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use TheBloke/dolphin-2.5-mixtral-8x7b-GPTQ with Docker Model Runner:
```
docker model run hf.co/TheBloke/dolphin-2.5-mixtral-8x7b-GPTQ
```

Malware api call sequences embeddings

by Akimbofmg9 - opened Dec 19, 2023

Discussion

Akimbofmg9

Dec 19, 2023

Hello,

I have a list of api call sequences from emulated malware logs, I want to classify the call sequences with their arguments, so I need embeddings. I chose this as this model can code.

I plan on using these embeddings in a Graph transformer, but my question is, how do i query get the model to generate embeddings for me.

Im using the feature extraction pipeline, but the issue is the context length, I have strings of API calls more than 3k in count. How would I make it work?
I'm pretty new to this so please dumb it down a little.
Any clues?

rpeinl

Dec 20, 2023

•

edited Dec 20, 2023

I'm trying to run the Mixtral models like dolphin here with the provided code, but it seems that it is not yet supported in transformers. I upgraded to the latest version as suggested (transformers-4.37.0.dev0) but it still gives me the following error
File /opt/conda/lib/python3.10/site-packages/transformers/models/auto/auto_factory.py:566,
ValueError(
570 f"Unrecognized configuration class {config.class} for this kind of AutoModel: {cls.name}.\n"
571 f"Model type should be one of {', '.join(c.name for c in cls._model_mapping.keys())}."
572 )
...
/opt/conda/lib/python3.10/site-packages/auto_gptq/nn_modules/qlinear/qlinear_exllama.py:68, in QuantLinear.init(self, bits, group_size, infeatures, outfeatures, bias, trainable, **kwargs)
66 assert infeatures % 32 == 0
67 assert infeatures % self.group_size == 0
---> 68 assert outfeatures % 32 == 0
70 self.register_buffer(
71 'qweight',
72 torch.zeros((infeatures // 32 * self.bits, outfeatures), dtype=torch.int32)
73 )
74 self.register_buffer(
75 'qzeros',
76 torch.zeros((math.ceil(infeatures / self.group_size), outfeatures // 32 * self.bits), dtype=torch.int32)
77 )

AssertionError:

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