Instructions to use roneneldan/TinyStories-33M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use roneneldan/TinyStories-33M with Transformers:

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

pipe = pipeline("text-generation", model="roneneldan/TinyStories-33M")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("roneneldan/TinyStories-33M")
model = AutoModelForCausalLM.from_pretrained("roneneldan/TinyStories-33M")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use roneneldan/TinyStories-33M with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "roneneldan/TinyStories-33M"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "roneneldan/TinyStories-33M",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/roneneldan/TinyStories-33M

SGLang

How to use roneneldan/TinyStories-33M 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 "roneneldan/TinyStories-33M" \
    --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": "roneneldan/TinyStories-33M",
		"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 "roneneldan/TinyStories-33M" \
        --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": "roneneldan/TinyStories-33M",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use roneneldan/TinyStories-33M with Docker Model Runner:
```
docker model run hf.co/roneneldan/TinyStories-33M
```

How long was this model trained?

by jploski - opened Sep 2, 2023

Discussion

jploski

Sep 2, 2023

How many steps/epochs was this particular model trained on? And which of the datasets was used: was it https://huggingface.co/datasets/roneneldan/TinyStories/tree/main/TinyStories-train.txt?

I can only find Figure 3 in the paper showing 2.5K steps. Am I right that it translates into ~1.5 epochs using the TinyStories-train.txt dataset and parameters from the model card?

roneneldan

Owner Sep 2, 2023

•

edited Sep 2, 2023

About 20 epochs. Context length 512, batch size 80 (20 per device over 4 V-100 GPUs), 16 gradient accumulation steps. Learning rate 5e-4, wd=0.1, betas 0.9,0.95. The file used to train was indeed https://huggingface.co/datasets/roneneldan/TinyStories/blob/main/TinyStories-train.txt.

jploski

Sep 2, 2023

Thanks for the detailed info!

jploski changed discussion status to closed Sep 2, 2023

alexkstern

Aug 11, 2025

Hi @roneneldan . I wanted to ask, are these the same training details for the 1M param model?

Matheart

Oct 22, 2025

•

edited Oct 22, 2025

So is this 33M model trained using 4 GPUs within 30 hours? Or it can be trained in 30 hours by a single GPU?

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