Instructions to use LeoLM/leo-hessianai-70b-chat with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use LeoLM/leo-hessianai-70b-chat with Transformers:

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

pipe = pipeline("text-generation", model="LeoLM/leo-hessianai-70b-chat")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("LeoLM/leo-hessianai-70b-chat")
model = AutoModelForCausalLM.from_pretrained("LeoLM/leo-hessianai-70b-chat", 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 LeoLM/leo-hessianai-70b-chat with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "LeoLM/leo-hessianai-70b-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-hessianai-70b-chat",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/LeoLM/leo-hessianai-70b-chat

SGLang

How to use LeoLM/leo-hessianai-70b-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-hessianai-70b-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-hessianai-70b-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-hessianai-70b-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-hessianai-70b-chat",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use LeoLM/leo-hessianai-70b-chat with Docker Model Runner:
```
docker model run hf.co/LeoLM/leo-hessianai-70b-chat
```

AWQ-Variante

by SebastianBodza - opened Dec 6, 2023

Discussion

SebastianBodza

Dec 6, 2023

Hallo,

wäre es möglich, dass Ihr eine AWQ Variante bereitstellt?

Danke schonmal

McNau

Dec 7, 2023

GGUF would also be nice :)

bjoernp

LAION LeoLM org Dec 7, 2023

@TheBloke Can you please add this model (LeoLM/leo-hessianai-70b-chat) to your quantization queue? :)

bjoernp

LAION LeoLM org Dec 7, 2023

For now, you guys can also check out DiscoResearch/DiscoLM-70b, which is the Leo-70b model trained on mostly English instruction data, but it still speaks German very well. There are already quantization available for this.

McNau

Dec 9, 2023

For now, you guys can also check out DiscoResearch/DiscoLM-70b, which is the Leo-70b model trained on mostly English instruction data, but it still speaks German very well. There are already quantization available for this.

DiscoLM exhibits very strange behavior in German. Many words are duplicated. I cannot understand why. It seems like something goes wrong with token sampling.

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