Instructions to use Adi-cookieNmilk/writing-tools-qwen3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Adi-cookieNmilk/writing-tools-qwen3 with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-4B-Instruct-2507")
model = PeftModel.from_pretrained(base_model, "Adi-cookieNmilk/writing-tools-qwen3")

Transformers

How to use Adi-cookieNmilk/writing-tools-qwen3 with Transformers:

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

pipe = pipeline("text-generation", model="Adi-cookieNmilk/writing-tools-qwen3")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("Adi-cookieNmilk/writing-tools-qwen3", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Adi-cookieNmilk/writing-tools-qwen3 with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/Adi-cookieNmilk/writing-tools-qwen3

SGLang

How to use Adi-cookieNmilk/writing-tools-qwen3 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 "Adi-cookieNmilk/writing-tools-qwen3" \
    --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": "Adi-cookieNmilk/writing-tools-qwen3",
		"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 "Adi-cookieNmilk/writing-tools-qwen3" \
        --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": "Adi-cookieNmilk/writing-tools-qwen3",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Adi-cookieNmilk/writing-tools-qwen3 with Docker Model Runner:
```
docker model run hf.co/Adi-cookieNmilk/writing-tools-qwen3
```

writing-tools-qwen3

LoRA adapter for Qwen3-4B-Instruct-2507, fine-tuned for three Apple Intelligence–style writing tasks:

Summarise — concise paragraph summaries
Rewrite — formal/casual tone rewrites
Smart Reply — email replies as JSON (subject, body, tone)

Approach

QLoRA 4-bit NF4 fine-tuning (r=16, α=32) on ~3.6k SFT examples. Part of writing-tools-nlp, which also implements speculative decoding, guided generation, and MLX on-device inference.

Evaluation (val split, n=400)

Metric	Base Qwen3-4B	Fine-tuned
ROUGE-1	0.3877	0.5359
ROUGE-2	0.1905	0.3788
ROUGE-L	0.3139	0.4838
BERTScore-F1	0.8756	0.9136

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

base = "Qwen/Qwen3-4B-Instruct-2507"
model = AutoModelForCausalLM.from_pretrained(base, device_map="auto")
model = PeftModel.from_pretrained(model, "aditya-ice/writing-tools-qwen3")
tokenizer = AutoTokenizer.from_pretrained(base)

messages = [
    {"role": "system", "content": "You are a writing assistant. Summarise the following text in one concise paragraph."},
    {"role": "user", "content": "Your text here..."},
]
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True, enable_thinking=False)
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Training

Base model: Qwen/Qwen3-4B-Instruct-2507
Method: QLoRA (peft + trl SFTTrainer)
Data: CNN/DailyMail, PAWS, Enron emails (~4k examples)
Hardware: Colab T4, 1 epoch

Downloads last month: 19

Model tree for Adi-cookieNmilk/writing-tools-qwen3

Base model

Qwen/Qwen3-4B-Instruct-2507

Adapter

(5598)

this model