Instructions to use LeoLM/leo-mistral-hessianai-7b-chat with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use LeoLM/leo-mistral-hessianai-7b-chat with Transformers:

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

pipe = pipeline("text-generation", model="LeoLM/leo-mistral-hessianai-7b-chat")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("LeoLM/leo-mistral-hessianai-7b-chat")
model = AutoModelForMultimodalLM.from_pretrained("LeoLM/leo-mistral-hessianai-7b-chat")
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]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use LeoLM/leo-mistral-hessianai-7b-chat with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "LeoLM/leo-mistral-hessianai-7b-chat"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LeoLM/leo-mistral-hessianai-7b-chat",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/LeoLM/leo-mistral-hessianai-7b-chat

SGLang

How to use LeoLM/leo-mistral-hessianai-7b-chat 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 "LeoLM/leo-mistral-hessianai-7b-chat" \
    --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": "LeoLM/leo-mistral-hessianai-7b-chat",
		"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 "LeoLM/leo-mistral-hessianai-7b-chat" \
        --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": "LeoLM/leo-mistral-hessianai-7b-chat",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use LeoLM/leo-mistral-hessianai-7b-chat with Docker Model Runner:
```
docker model run hf.co/LeoLM/leo-mistral-hessianai-7b-chat
```

Use model for RAG application

by Lue-C - opened Jan 12, 2024

Discussion

Lue-C

Jan 12, 2024

Hi there,

I want to use the model for a RAG application. It works very well getting the relevant information and fusing it into an answer. But there is one essential problem I encounter:
When I ask a question which is not covered by the given information, It creates an answer although I asked it not to. I tried several prompts, like

prompt_template = """
    <|im_start|>system
    Beantworte die gestellte Frage. Benutze dabei ausschließlich folgende Informationen und nicht dein internes Wissen. Nutze lediglich die Informationen, die zur Beantwortung der Frage notwendig sind und gebe die Metadaten dieser Informationen an. 
    Wenn die Informationen nicht ausreichen um die Frage zu beantworten, dann sage, dass die Informationen nicht ausreichen.
    {context}
    <|im_end|>
    <|im_start|>user
    {question}<|im_end|>
    <|im_start|>assistant
    """

So, my question is the following: how can I get the model to deny an answer if the question is not covered by the given information?

sohrabt

Apr 11, 2024

Check out this chapter https://github.com/jphme/EM_German?tab=readme-ov-file#factual-retrieval--rag in the README.

Stefan-LTB

Apr 11, 2024

Hi,
I am also in the process of developing a RAG application for german data during my internship semester.
Can I ask what Framework and Tokenizer, Embedding modell you use?
Till now I experimented with custom models in ollama and looked a bit into langchain and haystack.

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