Instructions to use h2oai/h2ogpt-4096-llama2-70b-chat with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use h2oai/h2ogpt-4096-llama2-70b-chat with Transformers:

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

pipe = pipeline("text-generation", model="h2oai/h2ogpt-4096-llama2-70b-chat")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("h2oai/h2ogpt-4096-llama2-70b-chat")
model = AutoModelForCausalLM.from_pretrained("h2oai/h2ogpt-4096-llama2-70b-chat", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use h2oai/h2ogpt-4096-llama2-70b-chat with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "h2oai/h2ogpt-4096-llama2-70b-chat"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "h2oai/h2ogpt-4096-llama2-70b-chat",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/h2oai/h2ogpt-4096-llama2-70b-chat

SGLang

How to use h2oai/h2ogpt-4096-llama2-70b-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 "h2oai/h2ogpt-4096-llama2-70b-chat" \
    --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": "h2oai/h2ogpt-4096-llama2-70b-chat",
		"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 "h2oai/h2ogpt-4096-llama2-70b-chat" \
        --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": "h2oai/h2ogpt-4096-llama2-70b-chat",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use h2oai/h2ogpt-4096-llama2-70b-chat with Docker Model Runner:
```
docker model run hf.co/h2oai/h2ogpt-4096-llama2-70b-chat
```

Take very long time to get the answers and answers never comes

by UDJA - opened Jan 5, 2024

Discussion

UDJA

Jan 5, 2024

I installed h2ogpt in a VM and It has 64GB RAM and 750GB storage and intel Xeon processor. I tried with this model. but when I asked something It takes very long time to process more than 5,6 hours and no any result. please help.

pseudotensor

H2O.ai org Jan 5, 2024

70b without GPU on FP16 isn't good idea. Use GGUF model and use zephyr 7b beta or open_chat.

pseudotensor changed discussion status to closed Jan 5, 2024

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