Llama
Meta's open-weight large language model family, the dominant open-source alternative to GPT and Claude.
Full Definition
Llama (Large Language Model Meta AI) is a family of open-weight language models released by Meta AI, starting with Llama 1 in February 2023. The release of Llama weights to researchers sparked an explosion of open-source AI development, enabling fine-tuning, quantisation, and deployment on consumer hardware. Llama 3, released in 2024 with 8B and 70B parameter variants, is competitive with GPT-3.5 and approaches GPT-4 on many benchmarks. Because the weights are publicly available, Llama has become the base for hundreds of fine-tuned derivatives (Mistral, Vicuna, WizardLM, etc.). 'Open-weight' is more precise than 'open-source' since the training data is not released.
Examples
Running Llama 3 8B locally using Ollama on a MacBook Pro for privacy-sensitive document analysis.
Fine-tuning Llama 3 70B on a company's internal knowledge base to create a private enterprise assistant.
Apply this in your prompts
PromptITIN automatically uses techniques like Llama to build better prompts for you.
Related Terms
Open-Source Model
A model whose weights are publicly available, enabling local deployment, inspect…
View →Fine-Tuned Model
A pretrained model whose weights have been updated on a specific dataset for a t…
View →Large Language Model
A neural network with billions of parameters trained on text to understand and g…
View →