Llama

E472883

foundation model large language model family

Llama is an open-source large language model family developed by Meta for a wide range of natural language understanding and generation tasks.

Jump to: Statements Referenced by

Statements (51)

Predicate	Object
instanceOf	foundation model ⓘ large language model family ⓘ
capability	code generation ⓘ natural language generation ⓘ natural language understanding ⓘ question answering ⓘ summarization ⓘ text generation ⓘ translation ⓘ
developer	Meta NERFINISHED ⓘ Meta AI NERFINISHED ⓘ
distributionForm	inference APIs ⓘ model weights ⓘ
field	artificial intelligence ⓘ natural language processing ⓘ
hasSubModel	Llama 2 13B NERFINISHED ⓘ Llama 2 70B NERFINISHED ⓘ Llama 2 7B NERFINISHED ⓘ Llama 3 70B NERFINISHED ⓘ Llama 3 8B NERFINISHED ⓘ Llama 3.1 405B NERFINISHED ⓘ Llama 3.1 70B NERFINISHED ⓘ Llama 3.1 8B NERFINISHED ⓘ Llama 3.2 11B NERFINISHED ⓘ Llama 3.2 1B NERFINISHED ⓘ Llama 3.2 3B NERFINISHED ⓘ Llama 3.2 90B NERFINISHED ⓘ
hasVersion	Llama 1 NERFINISHED ⓘ Llama 2 NERFINISHED ⓘ Llama 3 NERFINISHED ⓘ Llama 3.1 NERFINISHED ⓘ Llama 3.2 NERFINISHED ⓘ
intendedUse	commercial applications ⓘ research ⓘ
licenseModel	open-source license ⓘ
notableFor	open availability of high-parameter LLMs ⓘ
optimization	instruction tuning ⓘ reinforcement learning from human feedback ⓘ
organization	Meta Platforms, Inc. NERFINISHED ⓘ
softwareType	open-source software ⓘ
supportsTask	chatbot applications ⓘ content creation ⓘ research assistance ⓘ software development assistance ⓘ virtual assistants ⓘ
trainingDataType	articles ⓘ books ⓘ code ⓘ web text ⓘ
trainingMethod	self-supervised learning ⓘ transformer architecture ⓘ

Full triples — surface form annotated when it differs from this entity's canonical label.

Meta AI assistant → usesModelFamily → Llama ⓘ