Instructions to use openchat/openchat-3.5-1210 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use openchat/openchat-3.5-1210 with Transformers:

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

pipe = pipeline("text-generation", model="openchat/openchat-3.5-1210")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("openchat/openchat-3.5-1210")
model = AutoModelForCausalLM.from_pretrained("openchat/openchat-3.5-1210")
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

vLLM

How to use openchat/openchat-3.5-1210 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "openchat/openchat-3.5-1210"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "openchat/openchat-3.5-1210",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/openchat/openchat-3.5-1210

SGLang

How to use openchat/openchat-3.5-1210 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 "openchat/openchat-3.5-1210" \
    --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": "openchat/openchat-3.5-1210",
		"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 "openchat/openchat-3.5-1210" \
        --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": "openchat/openchat-3.5-1210",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use openchat/openchat-3.5-1210 with Docker Model Runner:
```
docker model run hf.co/openchat/openchat-3.5-1210
```

What is the base model of openchat ? Llama /mistral / custom ?

by StephanePop - opened Dec 20, 2023

Discussion

StephanePop

Dec 20, 2023

Hello,

I'm interested in using OpenChat 3.5 for commercial applications, given its Apache 2.0 license.
However, I've observed that previous versions of OpenChat were developed on the Llama2 framework.

Does OpenChat 3.5 also rely on the Llama2 base?
Additionally, does the training mentioned on the model's page refer primarily to fine-tuning?

Thank you in advance for your answer

imone

OpenChat org Dec 20, 2023

This model is based on Mistral 7B (Apache-2.0 licensed). Besides, the training mentioned refers to fine-tuning.

EladC

Dec 21, 2023

I would like to uderstand one thing, the paper only mention a 13b model with llama-2-13b as its base model.
so if i understand correctly:

openchat-13b (from the paper) -> base model llama-2-13b
openchat 3.5 (this repo) -> base model Mistral 7b

imone

OpenChat org Dec 21, 2023

Yes, exactly. This release is based on Mistral 7b.

EladC

Dec 21, 2023

Alright, thank you for the great work.

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