Instructions to use windprak/open_steuerllm with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use windprak/open_steuerllm with Transformers:

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("windprak/open_steuerllm")
model = AutoModelForCausalLM.from_pretrained("windprak/open_steuerllm")
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 windprak/open_steuerllm with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/windprak/open_steuerllm

SGLang

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

Docker Model Runner
How to use windprak/open_steuerllm with Docker Model Runner:
```
docker model run hf.co/windprak/open_steuerllm
```

Improve model card with metadata and links

by nielsr HF Staff - opened Feb 12

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+45

-5

Files changed (1) hide show

README.md +45 -5

README.md CHANGED Viewed

@@ -1,13 +1,49 @@
-We present SteuerLLM, a domain-adapted LLM for German tax law trained on a large-scale synthetic dataset generated from authentic examination material using a controlled retrieval-augmented pipeline. SteuerLLM (28B parameters) consistently outperforms general-purpose instruction-tuned models of comparable size and, in several cases, substantially larger systems, demonstrating that domain-specific data and architectural adaptation are more decisive than parameter scale for performance on realistic legal reasoning tasks.
-Demo: https://steuerllm.i5.ai.fau.de/
-Checkout all the details on Github: https://github.com/windprak/steuerllm
-It can be served via Transformers, vLLM, SGlang, .. basically any Framework since it is based on Mistral (Expanded Mistral Small from 24B to 28B).
-Recommended Temperature: 0.3
 ```bibtex
 @article{steuerllm,
@@ -18,3 +54,7 @@ Recommended Temperature: 0.3
   url = {https://arxiv.org/abs/2602.11081}
 }
 ```

+---
+language:
+- de
+license: other
+library_name: transformers
+pipeline_tag: text-generation
+datasets:
+- windprak/steuerllm_pretraining_dataset
+- windprak/steuerllm_instruct_dataset
+- windprak/SteuerEx
+---
+# SteuerLLM: Local specialized large language model for German tax law analysis
+SteuerLLM is a domain-adapted Large Language Model (LLM) consisting of 28 billion parameters, specifically designed for German tax law analysis. It was introduced in the paper [SteuerLLM: Local specialized large language model for German tax law analysis](https://huggingface.co/papers/2602.11081).
+The model excels in domains governed by strict formal rules, precise terminology, and legally binding structures, such as tax law, where correct answers require exact statutory citation, structured legal argumentation, and numerical accuracy.
+- **Paper:** [SteuerLLM: Local specialized large language model for German tax law analysis](https://huggingface.co/papers/2602.11081)
+- **GitHub Repository:** [https://github.com/windprak/steuerllm](https://github.com/windprak/steuerllm)
+- **Demo:** [https://steuerllm.i5.ai.fau.de/](https://steuerllm.i5.ai.fau.de/)
+## Model Description
+SteuerLLM is based on an expanded Mistral Small architecture (extended from 24B to 28B parameters through a block expansion method). It was trained on a large-scale synthetic dataset generated from authentic German university tax law examination material using a controlled retrieval-augmented pipeline.
+The training procedure follows a two-stage approach:
+1.  **Continual Pretraining:** The base model's representations are adapted to tax-specific terminology and concepts by pretraining on domain-filtered web data.
+2.  **Instruction Fine-tuning:** The model is then fine-tuned on synthetically generated question-answer pairs derived from primary German legal sources (e.g., EStG, AO, KStG) using the "Water Fountain Algorithm." This algorithm employs retrieval-augmented generation with semantic ranking to ensure factual grounding and contextual relevance.
+SteuerLLM consistently outperforms general-purpose instruction-tuned models of comparable size and, in several cases, substantially larger systems, demonstrating the critical role of domain-specific data and architectural adaptation for performance on realistic legal reasoning tasks.
+## Evaluation
+The model's performance was evaluated using **SteuerEx**, the first open benchmark derived from authentic German university tax law examinations. SteuerEx comprises 115 expert-validated examination questions spanning six core tax law domains and multiple academic levels, utilizing a statement-level, partial-credit evaluation framework.
+## Usage
+SteuerLLM can be served via various frameworks, including **Transformers**, **vLLM**, and **SGLang**, as it is based on the Mistral architecture.
+**Recommended Inference Parameters:**
+-   **Temperature:** 0.3
+## Citation
+If you use this work, please cite:
 ```bibtex
 @article{steuerllm,
   url = {https://arxiv.org/abs/2602.11081}
 }
 ```
+## License
+Research and academic use only.