Instructions to use Maykeye/TinyLLama-v0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Maykeye/TinyLLama-v0 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="Maykeye/TinyLLama-v0")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("Maykeye/TinyLLama-v0") model = AutoModelForCausalLM.from_pretrained("Maykeye/TinyLLama-v0") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use Maykeye/TinyLLama-v0 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Maykeye/TinyLLama-v0" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Maykeye/TinyLLama-v0", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/Maykeye/TinyLLama-v0
- SGLang
How to use Maykeye/TinyLLama-v0 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 "Maykeye/TinyLLama-v0" \ --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": "Maykeye/TinyLLama-v0", "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 "Maykeye/TinyLLama-v0" \ --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": "Maykeye/TinyLLama-v0", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use Maykeye/TinyLLama-v0 with Docker Model Runner:
docker model run hf.co/Maykeye/TinyLLama-v0
Thank you for creating this model!
Hi @Maykeye ,
Just wanted to say thanks for creating this model - it's really useful to have a small Llama that still generates reasonable text for testing purposes!
We've integrated it to test our project in GitHub CI, since it has tight RAM constraints that doesn't allow to fit even OpenLlama-3B.
You are welcome! I'm pleasently surprised my model found users tbh
Your model was also handy for testing llamafiles CICD via https://github.com/Mozilla-Ocho/llamafile/tree/main/models
FYI it's a gguf as I've converted it to gguf via https://huggingface.co/mofosyne/TinyLLama-v0-5M-F16-llamafile