Instructions to use Aleph-Alpha/Kolibri-1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Aleph-Alpha/Kolibri-1 with Transformers:

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

pipe = pipeline("text-generation", model="Aleph-Alpha/Kolibri-1")
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("Aleph-Alpha/Kolibri-1")
model = AutoModelForCausalLM.from_pretrained("Aleph-Alpha/Kolibri-1", 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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Aleph-Alpha/Kolibri-1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Aleph-Alpha/Kolibri-1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Aleph-Alpha/Kolibri-1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Aleph-Alpha/Kolibri-1

SGLang

How to use Aleph-Alpha/Kolibri-1 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 "Aleph-Alpha/Kolibri-1" \
    --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": "Aleph-Alpha/Kolibri-1",
		"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 "Aleph-Alpha/Kolibri-1" \
        --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": "Aleph-Alpha/Kolibri-1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Aleph-Alpha/Kolibri-1 with Docker Model Runner:
```
docker model run hf.co/Aleph-Alpha/Kolibri-1
```

[Evaluation Report] Empirical NIS-2 Stress Test: Closed-Book Drift vs. RAG Grounding on H200 (1,850 Runs)

by sbeierle - opened 23 days ago

Discussion

sbeierle

23 days ago

Hello everyone / @maintainers,

We conducted an empirical baseline robustness and statutory compliance evaluation on Kolibri-1 (78B MoE) deployed on an NVIDIA H200 instance using the vLLM runtime.

The benchmark suite consists of 1,850 execution passes focusing on German regulatory constraints (NIS-2 directive / BSIG), statutory deadlines, and structural prompt injections.

Key Empirical Metrics:

Adversarial Hardening: 97.5% (39/40) prompt injection vectors successfully deflected. Baseline resistance against system-prompt overrides is remarkably solid.
RAG Grounded Inference: 96.4% – 100% statutory accuracy when context anchors are present.
Closed-Book Retention: Under zero-context unconstrained decoding, factual recall for critical statutory deadlines (e.g., multi-tier reporting windows) dropped to ~40.0% – 53.0%, showing characteristic late-layer entropy drift.

Dataset & Artifacts:

The complete benchmark configuration, prompt datasets, and raw execution logs are published here:
👉 Dataset: https://huggingface.co/datasets/sbeierle/kolibri-nis2-compliance-eval

Sharing this data in case it provides useful baseline signals for ongoing post-training, alignment evaluations, or inference-time steering workflows.

sbeierle

23 days ago

sbeierle changed discussion status to closed 23 days ago

sbeierle changed discussion status to open 22 days ago

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