Instructions to use sail/Sailor-7B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use sail/Sailor-7B with Transformers:

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

pipe = pipeline("text-generation", model="sail/Sailor-7B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("sail/Sailor-7B")
model = AutoModelForCausalLM.from_pretrained("sail/Sailor-7B")
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

vLLM

How to use sail/Sailor-7B with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/sail/Sailor-7B

SGLang

How to use sail/Sailor-7B 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 "sail/Sailor-7B" \
    --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": "sail/Sailor-7B",
		"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 "sail/Sailor-7B" \
        --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": "sail/Sailor-7B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use sail/Sailor-7B with Docker Model Runner:
```
docker model run hf.co/sail/Sailor-7B
```

Any plan to open source the 200 B token dataset?

by YaoLiu61 - opened Apr 28, 2024

Discussion

YaoLiu61

Apr 28, 2024

Hi, thanks to your great work!
Do you have a plan to open source the 200 B token dataset used to continue pretrain the base model (Qwen1.5-7B)?

SivilTaram

Sea AI Lab org Apr 29, 2024

@YaoLiu61 Thanks for your interest! Sorry we cannot release the dataset due to several reasons (we have tried our best but there are still challenges). However, we will provide the full data cleaning code, and you may use the code to reproduce the dataset IMO.

YaoLiu61

Apr 29, 2024

@SivilTaram Thanks for your reply. And, when will the full data cleaning code be released ?

SivilTaram

Sea AI Lab org Apr 29, 2024

@YaoLiu61 tonight

SivilTaram

Sea AI Lab org Apr 30, 2024

@YaoLiu61 Hi, we just released the code for data cleaning / data deduplication at https://github.com/sail-sg/sailcraft

SivilTaram changed discussion status to closed May 4, 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