Instructions to use liyinghong/BioQwen-0.5B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use liyinghong/BioQwen-0.5B with Transformers:

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

pipe = pipeline("text-generation", model="liyinghong/BioQwen-0.5B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("liyinghong/BioQwen-0.5B")
model = AutoModelForCausalLM.from_pretrained("liyinghong/BioQwen-0.5B", device_map="auto")
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 liyinghong/BioQwen-0.5B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "liyinghong/BioQwen-0.5B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "liyinghong/BioQwen-0.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/liyinghong/BioQwen-0.5B

SGLang

How to use liyinghong/BioQwen-0.5B 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 "liyinghong/BioQwen-0.5B" \
    --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": "liyinghong/BioQwen-0.5B",
		"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 "liyinghong/BioQwen-0.5B" \
        --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": "liyinghong/BioQwen-0.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use liyinghong/BioQwen-0.5B with Docker Model Runner:
```
docker model run hf.co/liyinghong/BioQwen-0.5B
```

BioQwen-0.5B / README.md

yueqingyou

Create README.md

822467b verified about 2 years ago

preview code

Raw

History Blame Contribute Delete

2.56 kB

metadata

license: mit
datasets:
  - yueqingyou/BioQwen
language:
  - en
  - zh
tags:
  - BioQwen
  - 0.5B
  - Biomedical
  - Multi-Tasks

BioQwen: A Small-Parameter, High-Performance Bilingual Model for Biomedical Multi-Tasks

For model inference, please refer to the following example code:

import torch
import transformers
from transformers import AutoTokenizer, AutoModelForCausalLM

transformers.logging.set_verbosity_error()
max_length = 512
model_path = 'yueqingyou/BioQwen-0.5B'
tokenizer = AutoTokenizer.from_pretrained(model_path, use_fast=True)
model = AutoModelForCausalLM.from_pretrained(model_path, device_map='auto', torch_dtype=torch.bfloat16, attn_implementation='flash_attention_2').eval()

def predict(prompt):
    zh_system = "你是千问生物智能助手，一个专注于生物领域的先进人工智能。"
    en_system = "You are BioQwen, an advanced AI specializing in the field of biology."
    
    english_count, chinese_count = 0, 0
    for char in prompt:
        if '\u4e00' <= char <= '\u9fff':
            chinese_count += 1
        elif 'a' <= char.lower() <= 'z':
            english_count += 1
    lang = 'zh' if chinese_count > english_count else 'en'
    
    messages = [
        {"role": "system", "content": zh_system if lang == 'zh' else en_system},
        {"role": "user", "content": prompt}
    ]
    text = tokenizer.apply_chat_template(
        messages,
        tokenize=False,
        add_generation_prompt=True
    )

    model_inputs = tokenizer([text], return_tensors="pt").to('cuda')

    with torch.no_grad():
        generated_ids = model.generate(
            model_inputs.input_ids,
            max_new_tokens=max_length,
            eos_token_id=tokenizer.eos_token_id,
            pad_token_id=tokenizer.pad_token_id,
            do_sample=True,
            top_p = 0.9,
            temperature = 0.3,
            repetition_penalty = 1.1
        )

    generated_ids = [
        output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
    ]
    response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
    
    return response.strip()

prompt = 'I am suffering from irregular periods. I am currently taking medication Levothyroxine 50. My T3 is 0.87 ng/mL, T4 is 8.30 ug/dL, TSH is 2.43 uIU/mL. I am 34 years old, weigh 75 kg, and 5 feet tall. Please advice.'
print(f'Question:\t{prompt}\n\nAnswer:\t{predict(prompt)}')

For more detailed information and code, please refer to GitHub.