OpenELM 1 1B Instruct

OpenELM-1_1B-Instruct is a text generation model hosted on Hugging Face. It belongs to the class of causal language models implemented with the Transformers library and carries the apple-amlr license.

The model is provided by Apple and is associated with the arxiv paper 2404.14619. It uses a custom code implementation and is distributed in Safetensors format. Instructions for loading it appear on its repository page, which also shows community metrics such as 6.92k downloads and 75 likes.

Integration occurs through standard libraries. Code examples demonstrate use with the Transformers pipeline for text generation by specifying the model identifier and setting trust_remote_code to True. Direct loading employs AutoModelForCausalLM with the same parameters and automatic dtype selection. Support extends to vLLM through a pip-installed server started via a command-line instruction that accepts the model name, after which an OpenAI-compatible endpoint can be queried with curl. Notebooks for Google Colab and Kaggle are referenced as additional starting points.

The repository page supplies files and version history along with deployment options to inference providers or local applications. No pricing information is listed because the model is offered for download and local or server-based execution under its stated license.