Instructions to use miulab/Qwen3-1.7B-Usefulness with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use miulab/Qwen3-1.7B-Usefulness with Transformers:

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

pipe = pipeline("text-generation", model="miulab/Qwen3-1.7B-Usefulness")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("miulab/Qwen3-1.7B-Usefulness")
model = AutoModelForCausalLM.from_pretrained("miulab/Qwen3-1.7B-Usefulness", 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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use miulab/Qwen3-1.7B-Usefulness with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "miulab/Qwen3-1.7B-Usefulness"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "miulab/Qwen3-1.7B-Usefulness",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/miulab/Qwen3-1.7B-Usefulness

SGLang

How to use miulab/Qwen3-1.7B-Usefulness 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 "miulab/Qwen3-1.7B-Usefulness" \
    --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": "miulab/Qwen3-1.7B-Usefulness",
		"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 "miulab/Qwen3-1.7B-Usefulness" \
        --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": "miulab/Qwen3-1.7B-Usefulness",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use miulab/Qwen3-1.7B-Usefulness with Docker Model Runner:
```
docker model run hf.co/miulab/Qwen3-1.7B-Usefulness
```

theblackcat102 commited on Dec 15, 2025

Commit

647ff90

verified ·

1 Parent(s): 601605f

Update README.md

Browse files

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -17,10 +17,10 @@ We repeat the same evaluation for 10 times on our test set and report the averag
 | Model | Count | Mean | StdDev | CV% | Min | Max |
 |-------|-------|------|--------|-----|-----|-----|
 | Qwen3-1.7B (this) | 10 | 0.8248 | 0.0226 | 2.74 | 0.7815 | 0.8609 |
-| Qwen3-4B (private) | 10 | 0.8438 | 0.0116 | 1.38 | 0.8272 | 0.8608 |
 Direct Answer:
-| Metric | Qwen3-4B (private) | Qwen3-1.7B (this) |
 |--------|----------------------------------|-----------------------------------------------|
 | **Accuracy** | 0.8983 (89.83%) | 0.8644 (86.44%) |
 | **Precision** | 0.8205 (82.05%) | 0.8281 (82.81%) |

 | Model | Count | Mean | StdDev | CV% | Min | Max |
 |-------|-------|------|--------|-----|-----|-----|
 | Qwen3-1.7B (this) | 10 | 0.8248 | 0.0226 | 2.74 | 0.7815 | 0.8609 |
+| (private) | 10 | 0.8438 | 0.0116 | 1.38 | 0.8272 | 0.8608 |
 Direct Answer:
+| Metric | (private) | Qwen3-1.7B (this) |
 |--------|----------------------------------|-----------------------------------------------|
 | **Accuracy** | 0.8983 (89.83%) | 0.8644 (86.44%) |
 | **Precision** | 0.8205 (82.05%) | 0.8281 (82.81%) |