Instructions to use drooryck/babylm-macaroni with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use drooryck/babylm-macaroni with Transformers:

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

pipe = pipeline("text-generation", model="drooryck/babylm-macaroni")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("drooryck/babylm-macaroni")
model = AutoModelForCausalLM.from_pretrained("drooryck/babylm-macaroni", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use drooryck/babylm-macaroni with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "drooryck/babylm-macaroni"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "drooryck/babylm-macaroni",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/drooryck/babylm-macaroni

SGLang

How to use drooryck/babylm-macaroni 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 "drooryck/babylm-macaroni" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "drooryck/babylm-macaroni",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "drooryck/babylm-macaroni" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "drooryck/babylm-macaroni",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use drooryck/babylm-macaroni with Docker Model Runner:
```
docker model run hf.co/drooryck/babylm-macaroni
```

babylm-macaroni

A GPT-2 model for the BabyLM 2026 Multilingual track. It is trained with our code-switched curriculum corpus. This is our leaderboard submission with full checkpoints. Please see our full set of models for the paper here

Model Sources

The paper, repository, and further details are TBA on release.

Downloads last month: 430

Safetensors

Model size

98.4M params

Tensor type

F32