Instructions to use yi71/qwen3.5-4B-qlora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use yi71/qwen3.5-4B-qlora with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("unsloth/Qwen3.5-4B")
model = PeftModel.from_pretrained(base_model, "yi71/qwen3.5-4B-qlora")

Transformers

How to use yi71/qwen3.5-4B-qlora with Transformers:

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

pipe = pipeline("text-generation", model="yi71/qwen3.5-4B-qlora")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("yi71/qwen3.5-4B-qlora", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use yi71/qwen3.5-4B-qlora with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "yi71/qwen3.5-4B-qlora"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "yi71/qwen3.5-4B-qlora",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/yi71/qwen3.5-4B-qlora

SGLang

How to use yi71/qwen3.5-4B-qlora 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 "yi71/qwen3.5-4B-qlora" \
    --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": "yi71/qwen3.5-4B-qlora",
		"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 "yi71/qwen3.5-4B-qlora" \
        --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": "yi71/qwen3.5-4B-qlora",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Studio

How to use yi71/qwen3.5-4B-qlora with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for yi71/qwen3.5-4B-qlora to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for yi71/qwen3.5-4B-qlora to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for yi71/qwen3.5-4B-qlora to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="yi71/qwen3.5-4B-qlora",
    max_seq_length=2048,
)

Docker Model Runner
How to use yi71/qwen3.5-4B-qlora with Docker Model Runner:
```
docker model run hf.co/yi71/qwen3.5-4B-qlora
```

Qwen3.5-4B QLoRA Adapter

A QLoRA fine-tuned adapter for unsloth/Qwen3.5-4B, optimized for specific domain tasks.

Model Details

Base Model: unsloth/Qwen3.5-4B
Method: QLoRA (Quantized Low-Rank Adaptation)
Task: Causal Language Modeling (Text Generation)
Language: Chinese
Framework: PEFT + Transformers + TRL + Unsloth
License: Apache 2.0 (base model)

How to Use

Load with PEFT

from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

# Load base model
base_model = AutoModelForCausalLM.from_pretrained(
    "unsloth/Qwen3.5-4B",
    torch_dtype="auto",
    device_map="auto",
)

# Load adapter
model = PeftModel.from_pretrained(base_model, "yi71/qwen3.5-4B-qlora")
tokenizer = AutoTokenizer.from_pretrained("yi71/qwen3.5-4B-qlora")

# Inference
messages = [
    {"role": "user", "content": "你好，请介绍一下你自己。"}
]
text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(text, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=512)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Merge & Save

# Merge adapter into base model for standalone use
merged_model = model.merge_and_unload()
merged_model.save_pretrained("merged_model")
tokenizer.save_pretrained("merged_model")

Training Details

Parameter	Value
Method	QLoRA (4-bit quantization + LoRA)
Rank (r)	8
LoRA Alpha	16
LoRA Dropout	0.05
Target Modules	q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj
Quantization	4-bit (QLoRA)
Hardware	Single consumer GPU
PEFT Version	0.19.1

Adapter Config

{
  "peft_type": "LORA",
  "task_type": "CAUSAL_LM",
  "r": 8,
  "lora_alpha": 16,
  "lora_dropout": 0.05,
  "target_modules": ["q_proj", "k_proj", "v_proj", "o_proj", "gate_proj", "up_proj", "down_proj"]
}

Citation

@article{dettmers2023qlora,
  title={QLoRA: Efficient Finetuning of Quantized Language Models},
  author={Dettmers, Tim and Pagnoni, Artidoro and Holtzman, Ari and Zettlemoyer, Luke},
  journal={arXiv preprint arXiv:2305.14314},
  year={2023}
}

Framework Versions

PEFT 0.19.1
Transformers 4.51+
PyTorch 2.0+

Downloads last month: 2

Model tree for yi71/qwen3.5-4B-qlora

Base model

Qwen/Qwen3.5-4B-Base

Finetuned

Qwen/Qwen3.5-4B

Finetuned

unsloth/Qwen3.5-4B

Adapter

(46)

this model

Paper for yi71/qwen3.5-4B-qlora

QLoRA: Efficient Finetuning of Quantized LLMs

Paper • 2305.14314 • Published May 23, 2023 • 62