Instructions to use berkeley-nest/Starling-LM-7B-alpha with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use berkeley-nest/Starling-LM-7B-alpha with Transformers:

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

pipe = pipeline("text-generation", model="berkeley-nest/Starling-LM-7B-alpha")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("berkeley-nest/Starling-LM-7B-alpha")
model = AutoModelForCausalLM.from_pretrained("berkeley-nest/Starling-LM-7B-alpha")
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 berkeley-nest/Starling-LM-7B-alpha with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/berkeley-nest/Starling-LM-7B-alpha

SGLang

How to use berkeley-nest/Starling-LM-7B-alpha 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 "berkeley-nest/Starling-LM-7B-alpha" \
    --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": "berkeley-nest/Starling-LM-7B-alpha",
		"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 "berkeley-nest/Starling-LM-7B-alpha" \
        --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": "berkeley-nest/Starling-LM-7B-alpha",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use berkeley-nest/Starling-LM-7B-alpha with Docker Model Runner:
```
docker model run hf.co/berkeley-nest/Starling-LM-7B-alpha
```

Maybe you could try on Yi-34B

#12

by Yhyu13 - opened Nov 29, 2023

Discussion

Yhyu13

Nov 29, 2023

Hi,

Your base model was Openchat3.5 whose base was Mistral-7B, the sad thing is that they have not yet released larger models.

And from AlpacaEval https://tatsu-lab.github.io/alpaca_eval/, we generally observe the trend that larger models with same FT method perform better, e.g. the XWin series (which is also a RLHF model).

Maybe you can try out Yi-34B, seems to be the best mid-size model so far.

Thanks!

banghua

Berkeley-Nest org Nov 29, 2023

Thank you for the suggestion! That's also on our TO-DO list.

Currently we still observe some instable and weird behavior of the model, so we are working on a beta version first before testing a larger reward & policy model.

During our evaluation, we also found that 7B model tends to hallucilate a lot, which is incomparable to 30+B models and greatly affect the human evaluation score. So having a larger model seems to be a must in this case. We believe that our dataset might have larger potential when scaling the reward model and language model, although the biggest problem is still the limited compute for training large reward & language model.

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