Instructions to use thundax/Qwen2-1.5B-Sign with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use thundax/Qwen2-1.5B-Sign with Transformers:

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

pipe = pipeline("text-generation", model="thundax/Qwen2-1.5B-Sign")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("thundax/Qwen2-1.5B-Sign")
model = AutoModelForCausalLM.from_pretrained("thundax/Qwen2-1.5B-Sign")
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 thundax/Qwen2-1.5B-Sign with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "thundax/Qwen2-1.5B-Sign"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "thundax/Qwen2-1.5B-Sign",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/thundax/Qwen2-1.5B-Sign

SGLang

How to use thundax/Qwen2-1.5B-Sign 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 "thundax/Qwen2-1.5B-Sign" \
    --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": "thundax/Qwen2-1.5B-Sign",
		"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 "thundax/Qwen2-1.5B-Sign" \
        --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": "thundax/Qwen2-1.5B-Sign",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use thundax/Qwen2-1.5B-Sign with Docker Model Runner:
```
docker model run hf.co/thundax/Qwen2-1.5B-Sign
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

language:

en pipeline_tag: text-generation tags:

Qwen2-1.5B-Sign

Introduction

Qwen2-Sign is a text to sige model base on Qwen2.

Finetune Details

Finetune dataset: alpaca-zh-text2sign
Finetune parameter

Parameter	Value
learning_rate	5e-05
train_batch_size	4
eval_batch_size	4
gradient_accumulation_steps	8
total_train_batch_size	32
lr_scheduler_type	cosine
lr_scheduler_warmup_steps	100
num_epochs	4

Quickstart

from transformers import AutoModelForCausalLM, AutoTokenizer

device = "cuda"  # the device to load the model onto

model = AutoModelForCausalLM.from_pretrained(
    "thundax/Qwen2-1.5B-Sign",
    torch_dtype="auto",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained("thundax/Qwen2-1.5B-Sign")

text = "你好，世界！"
text = f'Translate sentence into labels\n{text}\n'
model_inputs = tokenizer([text], return_tensors="pt").to(device)

generated_ids = model.generate(
    model_inputs.input_ids,
    max_new_tokens=512
)
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]

Citation

If you find our work helpful, feel free to give us a cite.

@software{qwen2-sign,
  author = {thundax},
  title = {qwen2-sign: A Tool for Text to Sign},
  year = {2024},
  url = {https://github.com/thundax-lyp},
}

Downloads last month: 3

Safetensors

Model size

2B params

Tensor type

BF16

Model tree for thundax/Qwen2-1.5B-Sign

Quantizations

2 models

Collection including thundax/Qwen2-1.5B-Sign

Qwen2-Sign

Collection

1 item • Updated Mar 31, 2025