Instructions to use openchat/openchat-3.5-0106-gemma with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use openchat/openchat-3.5-0106-gemma with Transformers:

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

pipe = pipeline("text-generation", model="openchat/openchat-3.5-0106-gemma")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("openchat/openchat-3.5-0106-gemma")
model = AutoModelForCausalLM.from_pretrained("openchat/openchat-3.5-0106-gemma")
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 openchat/openchat-3.5-0106-gemma with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "openchat/openchat-3.5-0106-gemma"
# 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-3.5-0106-gemma",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/openchat/openchat-3.5-0106-gemma

SGLang

How to use openchat/openchat-3.5-0106-gemma 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-3.5-0106-gemma" \
    --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-3.5-0106-gemma",
		"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-3.5-0106-gemma" \
        --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-3.5-0106-gemma",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use openchat/openchat-3.5-0106-gemma with Docker Model Runner:
```
docker model run hf.co/openchat/openchat-3.5-0106-gemma
```

Wrong chat template?

by jordio - opened Mar 16, 2024

Discussion

jordio

Mar 16, 2024

I noticed the chat template for the new openchat-3.5-0106-gemma is not the one you list in the main page:

GPT4 Correct User: HelloGPT4 Correct Assistant: HiGPT4 Correct User: How are you today?GPT4 Correct Assistant:

Because if you check the "chat_template" key at:
https://huggingface.co/openchat/openchat-3.5-0106-gemma/blob/main/tokenizer_config.json

It's missing the "GPT4 Correct ..." prefixes.

So, which template is the correct? The one with "GPT4 Correct ..." or the one listed in the tokenizer_config?

(Thanks - the OpenChat models are great!)

Wanfq

Mar 19, 2024

I just updated "chat_template" in tokenizer_config.json with the following code: {{ bos_token }}{% for message in messages %}{{ 'GPT4 Correct ' + message['role'].title() + ': ' + message['content'] + '<end_of_turn>'}}{% endfor %}{% if add_generation_prompt %}{{ 'GPT4 Correct Assistant:' }}{% endif %}. As a result, the chat template on the main page now matches.

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