Instructions to use LevArtesa/grpo-humanizer-de with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use LevArtesa/grpo-humanizer-de with Transformers:

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

pipe = pipeline("text-generation", model="LevArtesa/grpo-humanizer-de")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("LevArtesa/grpo-humanizer-de")
model = AutoModelForCausalLM.from_pretrained("LevArtesa/grpo-humanizer-de")
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 LevArtesa/grpo-humanizer-de with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/LevArtesa/grpo-humanizer-de

SGLang

How to use LevArtesa/grpo-humanizer-de 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 "LevArtesa/grpo-humanizer-de" \
    --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": "LevArtesa/grpo-humanizer-de",
		"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 "LevArtesa/grpo-humanizer-de" \
        --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": "LevArtesa/grpo-humanizer-de",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use LevArtesa/grpo-humanizer-de with Docker Model Runner:
```
docker model run hf.co/LevArtesa/grpo-humanizer-de
```

GRPO Humanizer DE

Fine-tuned with Group Relative Policy Optimization (GRPO) to rewrite AI-generated German academic text so that it passes GPTZero detection while preserving semantic content.

Training details

Parameter	Value
Base model	Qwen/Qwen3-8B
Method	GRPO (TRL) + LoRA
Learning rate	5e-06
Batch size	2
Gradient accumulation	8
Max steps	50
Precision	bf16

Intended use

Academic text humanisation for German-language content. The model is designed to be called via the HuggingFace Inference API from the GhostWriter application.

Licence

Apache-2.0

Downloads last month: 198

Safetensors

Model size

8B params

Tensor type

F32

Model tree for LevArtesa/grpo-humanizer-de

Base model

Qwen/Qwen3-8B-Base

Finetuned

Qwen/Qwen3-8B

Adapter

(1436)

this model