Instructions to use Ace-2504/slm-125m-extended with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Ace-2504/slm-125m-extended with Transformers:

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

pipe = pipeline("text-generation", model="Ace-2504/slm-125m-extended")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Ace-2504/slm-125m-extended")
model = AutoModelForCausalLM.from_pretrained("Ace-2504/slm-125m-extended", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Ace-2504/slm-125m-extended with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Ace-2504/slm-125m-extended"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Ace-2504/slm-125m-extended",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Ace-2504/slm-125m-extended

SGLang

How to use Ace-2504/slm-125m-extended 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 "Ace-2504/slm-125m-extended" \
    --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": "Ace-2504/slm-125m-extended",
		"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 "Ace-2504/slm-125m-extended" \
        --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": "Ace-2504/slm-125m-extended",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Ace-2504/slm-125m-extended with Docker Model Runner:
```
docker model run hf.co/Ace-2504/slm-125m-extended
```

SLM-125M-extended — continued pretraining of slm-125m-base

Ace-2504/slm-125m-base continued for one epoch on ~515M additional tokens (389M FineWeb-Edu + 131M US case law) that the base model had never seen.

Constant learning rate 1e-4, optimizer state reset, 982 steps on one A100-40GB.

This model exists to answer a research question: does continued pretraining on new data improve a small model overall, or does it cause catastrophic forgetting?

Measured against the base model on its own held-out data, forgetting tracked each domain's share of the new data almost exactly:

Domain	share of new data	loss change
SEC filings	0%	+8.1% (worst)
US case law	25%	+1.0%
FineWeb-Edu	75%	-1.7% (improved)

Use Ace-2504/slm-125m-base if you want the original. Neither model is instruction-tuned; both are base models.

Downloads last month: 31

Safetensors

Model size

0.1B params

Tensor type

F32