Instructions to use allenai/OLMo-1B-hf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use allenai/OLMo-1B-hf with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="allenai/OLMo-1B-hf")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("allenai/OLMo-1B-hf") model = AutoModelForCausalLM.from_pretrained("allenai/OLMo-1B-hf") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use allenai/OLMo-1B-hf with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "allenai/OLMo-1B-hf" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "allenai/OLMo-1B-hf", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/allenai/OLMo-1B-hf
- SGLang
How to use allenai/OLMo-1B-hf 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 "allenai/OLMo-1B-hf" \ --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": "allenai/OLMo-1B-hf", "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 "allenai/OLMo-1B-hf" \ --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": "allenai/OLMo-1B-hf", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use allenai/OLMo-1B-hf with Docker Model Runner:
docker model run hf.co/allenai/OLMo-1B-hf
Correct model architecture name to match allenai/OLMo-1B
#2
by PierreLepagnol - opened
In this file: https://huggingface.co/allenai/OLMo-1B/blob/main/config.json the architecture is : "OLMoForCausalLM"
In this repo the architecture is : "OlmoForCausalLM"
This makes vllm pipeline fail.
Hi @PierreLepagnol , because this model (OLMo-1B-hf) is integrated directly with the transformers library (unlike its original model, OLMo-1B), the change in casing (from "OLMo" to "Olmo") was necessary for that integration.
Could you please detail the nature of the failure in the vllm pipeline you mentioned?
shanearora changed pull request status to closed