Instructions to use sileod/microlm-ettin-swa-5m with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use sileod/microlm-ettin-swa-5m with Transformers:

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

pipe = pipeline("text-generation", model="sileod/microlm-ettin-swa-5m")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("sileod/microlm-ettin-swa-5m")
model = AutoModelForCausalLM.from_pretrained("sileod/microlm-ettin-swa-5m")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use sileod/microlm-ettin-swa-5m with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "sileod/microlm-ettin-swa-5m"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "sileod/microlm-ettin-swa-5m",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/sileod/microlm-ettin-swa-5m

SGLang

How to use sileod/microlm-ettin-swa-5m 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 "sileod/microlm-ettin-swa-5m" \
    --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": "sileod/microlm-ettin-swa-5m",
		"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 "sileod/microlm-ettin-swa-5m" \
        --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": "sileod/microlm-ettin-swa-5m",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use sileod/microlm-ettin-swa-5m with Docker Model Runner:
```
docker model run hf.co/sileod/microlm-ettin-swa-5m
```

MicroLM-Ettin-SWA-5M

Tiny random-init causal LM for fast training smoke tests and pretraining pipeline dry runs.

This model is intentionally untrained. It is meant to test code paths, dataset packing, loss plumbing, checkpointing, evaluation loops, and distributed training setup without downloading or training a large model.

Specs

Architecture: ModernBertDecoderForCausalLM
Parameters: ~5M
Layers: 6
Hidden size: 256
Attention heads: 4
MLP intermediate size: 384
Context length: 512
Attention: sliding-window attention in every layer
Tokenizer: 4096-token tokenizer reused from SupraLabs/Supra-Mini-v4-2M
Weights: random initialization

Usage

from transformers import AutoTokenizer, AutoModelForCausalLM

repo_id = "sileod/microlm-ettin-swa-5m"

tok = AutoTokenizer.from_pretrained(repo_id)
model = AutoModelForCausalLM.from_pretrained(repo_id)

Downloads last month: 69

Safetensors

Model size

5.52M params

Tensor type

F32