Meta-Llama-3-8B-Instruct is an 8 billion parameter language model hosted on Hugging Face under the NousResearch organization. It belongs to the Llama 3 family and serves as an instruction-tuned variant intended for conversational and assistant-style applications.
The model uses the LlamaForCausalLM architecture with a model type of llama. Its tokenizer configuration includes a specific chat template that structures messages with role-based headers, beginning-of-text and end-of-text tokens, and support for an add-generation-prompt flag to prepare responses from an assistant role. The configuration also defines bos_token as <|begin_of_text| and eos_token as <|eot_id|.
It is delivered as a repository on the Hugging Face platform, where it can be accessed for download and use in machine learning workflows. The page indicates substantial community engagement through over two million all-time downloads. The model was created on April 18, 2024.
As a foundation model, it is positioned for research, fine-tuning, and integration into local AI systems that require instruction following. The repository is part of the broader Hugging Face ecosystem for open-source machine learning models.
Meta Llama 3 8B Instruct is a Foundation models & chat product. It focuses on providing an openly available, high-quality instruction-tuned language model for chat and assistant applications. Meta Llama 3 8B Instruct is an open-source project aimed at developers. The project is open source (Open Source). Meta Llama 3 8B Instruct is available on the web, the command line, and API.
NousResearch builds and maintains Meta Llama 3 8B Instruct, and the product first shipped in 2024. The GitHub repository has been archived. PulseGate's similarity index places it among 15 comparable tools. Among its 3 catalogued features are instruction tuning, chat template, and text generation.
Latest indexed changes and source events
NousResearch/Meta-Llama-3-8B-Instruct verified by the PulseGate indexer
⚠ Archived
Other apps tracked under the same category.
