Instructions to use malteos/german-r1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use malteos/german-r1 with Transformers:

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

pipe = pipeline("text-generation", model="malteos/german-r1")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("malteos/german-r1")
model = AutoModelForCausalLM.from_pretrained("malteos/german-r1")
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]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use malteos/german-r1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "malteos/german-r1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "malteos/german-r1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/malteos/german-r1

SGLang

How to use malteos/german-r1 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 "malteos/german-r1" \
    --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": "malteos/german-r1",
		"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 "malteos/german-r1" \
        --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": "malteos/german-r1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use malteos/german-r1 with Docker Model Runner:
```
docker model run hf.co/malteos/german-r1
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

German R1

Introducing German-R1. We are so back!!11

🇩🇪 German R1 is a reasoning model almost equivalent to OpenAI‘s o3 or DeepSeek‘s R1 - but it thinks in German!
🇩🇪 German R1 is so efficient that is was build without any government funding.
🇩🇪 German R1 was only trained with legally imported H100s in less than five minutes.
Please do NOT take this too serious.

Context

See Linkedin: https://www.linkedin.com/posts/activity-7294337496023269376-ZBkI

Usage

You can run the model as follows:

from transformers import pipeline, set_seed
import json

set_seed(42)
pipe = pipeline("text-generation", "malteos/german-r1")

# from gsm8k test set
question = "James beschließt, 3-mal pro Woche 3 Sprints zu laufen.  Er läuft 60 Meter pro Sprint.  Wie viele Meter läuft er insgesamt pro Woche?"
expected_answer = "540"

# xml reasoning and answer format
system_prompt = """
Antworte auf deutsch und in diesem Format:
<reasoning>
...
</reasoning>
<answer>
...
</answer>
"""

messages = [
    {
        "role": "system",
        "content": system_prompt,
    },
    {"role": "user", "content": dataset["question"][3]},
]
response = pipe(messages, max_new_tokens=256)

print(json.dumps(response, indent=4, ensure_ascii=False))

The output should look like this (correct answer: 540)

[
    {
        "generated_text": [
            {
                "role": "system",
                "content": "\nAntworte auf deutsch und in diesem Format:\n<reasoning>\n...\n</reasoning>\n<answer>\n...\n</answer>\n"
            },
            {
                "role": "user",
                "content": "James beschließt, 3-mal pro Woche 3 Sprints zu laufen.  Er läuft 60 Meter pro Sprint.  Wie viele Meter läuft er insgesamt pro Woche?"
            },
            {
                "role": "assistant",
                "content": "<reasoning>\nJames läuft 3 Sprints pro Tag, wobei jeder Sprint 60 Meter beträgt. Deshalb läuft er pro Tag 3 Sprints * 60 Meter/Sprint = 180 Meter. James läuft 3 Mal pro Woche, also läuft er insgesamt 180 Meter * 3 = 540 Meter pro Woche.\n</reasoning>\n<answer>\n540\n</answer>\n"
            }
        ]
    }
]

Training details

Base model: Qwen2.5-3B-Instruct
Implementation: based on GRPO demo by will brown + language identification reward
Training data: German subset of openGPT-X/gsm8kx (machine translated from the English gsm8k)