Instructions to use swiss-ai/Apertus-8B-Instruct-2509 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use swiss-ai/Apertus-8B-Instruct-2509 with Transformers:

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

pipe = pipeline("text-generation", model="swiss-ai/Apertus-8B-Instruct-2509")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("swiss-ai/Apertus-8B-Instruct-2509")
model = AutoModelForCausalLM.from_pretrained("swiss-ai/Apertus-8B-Instruct-2509", device_map="auto")
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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
HuggingChat
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use swiss-ai/Apertus-8B-Instruct-2509 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "swiss-ai/Apertus-8B-Instruct-2509"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "swiss-ai/Apertus-8B-Instruct-2509",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/swiss-ai/Apertus-8B-Instruct-2509

SGLang

How to use swiss-ai/Apertus-8B-Instruct-2509 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 "swiss-ai/Apertus-8B-Instruct-2509" \
    --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": "swiss-ai/Apertus-8B-Instruct-2509",
		"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 "swiss-ai/Apertus-8B-Instruct-2509" \
        --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": "swiss-ai/Apertus-8B-Instruct-2509",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use swiss-ai/Apertus-8B-Instruct-2509 with Docker Model Runner:
```
docker model run hf.co/swiss-ai/Apertus-8B-Instruct-2509
```

Suggestion: Sentence Transformer based on Apertus-8B-Instruct for Swiss GLAM, Memobase & Research

#14

by DanielMemoriav - opened Sep 8, 2025

Discussion

DanielMemoriav

Sep 8, 2025

Dear swiss-ai team,

Thank you very much for your impressive work and for making Apertus-8B-Instruct available to the community. This model is a valuable contribution to the Swiss AI ecosystem.

For many use cases in the swiss GLAM sector, in research, and specifically for Memobase, a dedicated Sentence Transformer based on your LLM would be extremely beneficial. Our goal is to enable robust semantic search in multilingual cultural heritage databases, which would directly support institutions like Memobase and similar projects.

Memobase is the Swiss aggregation platform for searching and accessing digitized cultural AV heritage operated by Memoriav, the association for the preservation of AV heritage in Switzerland.

Would you consider training and releasing a (compact) Sentence Transformer derived from Apertus-8B-Instruct as a community resource? This would greatly support Swiss GLAM organizations, Memobase, and the research community.

Thank you for your consideration and for your outstanding work!

Best regards,
Daniel

mjaggi

Swiss AI Initiative org Sep 22, 2025

yes it would be great if the community would help creating embedding models from apertus. this can be done for example by distillation and finetuning on an adjusted embedding objective.

here is some example by another contributor: https://huggingface.co/speakdatawith/Apertus-8B-2509-Encoder

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment