Instructions to use 01-ai/Yi-34B-Chat-4bits with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use 01-ai/Yi-34B-Chat-4bits with Transformers:

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

pipe = pipeline("text-generation", model="01-ai/Yi-34B-Chat-4bits")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("01-ai/Yi-34B-Chat-4bits")
model = AutoModelForCausalLM.from_pretrained("01-ai/Yi-34B-Chat-4bits")
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 Settings

vLLM

How to use 01-ai/Yi-34B-Chat-4bits with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "01-ai/Yi-34B-Chat-4bits"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "01-ai/Yi-34B-Chat-4bits",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/01-ai/Yi-34B-Chat-4bits

SGLang

How to use 01-ai/Yi-34B-Chat-4bits 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 "01-ai/Yi-34B-Chat-4bits" \
    --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": "01-ai/Yi-34B-Chat-4bits",
		"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 "01-ai/Yi-34B-Chat-4bits" \
        --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": "01-ai/Yi-34B-Chat-4bits",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use 01-ai/Yi-34B-Chat-4bits with Docker Model Runner:
```
docker model run hf.co/01-ai/Yi-34B-Chat-4bits
```

Can Yi-34B-Chat-4bits use vllm for inference?

by wangdafa - opened Nov 29, 2023

Discussion

wangdafa

Nov 29, 2023

The start command is as follows：
python -m fastchat.serve.vllm_worker --model-path 01-ai/Yi-34B-Chat-4bits --trust-remote-code --tensor-parallel-size 2 --quantization awq --max-model-len 4096 --model-name Qwen-14B-Chat
I found that the prompt template of Yi-34B-Chat is the same as Qwen-chat. When using vllm, you can specify the template by using the method of "--model-name Qwen-14B-Chat". If "--model-name Qwen-14B-Chat" is not added, the template will become the default one-shot of fastchat. However, starting the model in this way will cause the model to not output any content, although the model has already been loaded.

panyang01

01-ai org Nov 30, 2023

Please keep an eye on https://github.com/lm-sys/FastChat/pull/2723 to see if it can resolve your issue.

tianjun

01-ai org Dec 2, 2023

Please try the main branch of vllm.

tianjun changed discussion status to closed Dec 2, 2023

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