Skip to content

Get started with Llama

This guide provides information and resources to help you set up Llama including how to access the model, hosting, how-to and integration guides. Additionally, you will find supplemental materials to further assist you while building with Llama.

Llama models

Llama 4Natively multimodal models with a mixture-of-experts architecture.
Learn more
Llama 3The Llama 3 family, including 3.1, 3.2, and 3.3 across all instruction-tuned text models.
Learn more
Llama 2Open foundation and fine-tuned chat models for a wide variety of applications.
Learn more
Llama AI InnovationExplore the latest AI innovation use cases and applications powered by Llama models.
Learn more
Llama ProtectionsSafety tools, guardrails, and trust & safety resources for responsible AI deployment.
Learn more
AI DefendersProtect against AI-generated threats with Meta's defensive AI tools and resources.
Learn more

Llama license & use policy

Llama 4
Llama 4 LicenseLlama 4 Use policy
Llama 3
Llama 3.3 LicenseLlama 3.3 Use policyLlama 3.2 LicenseLlama 3.2 Use policyLlama 3.1 LicenseLlama 3.1 Use policyLlama 3 LicenseLlama 3 Use policy
Llama 2
Llama 2 LicenseLlama 2 Use policy

Llama documentation

Getting the Models
Overview1B/3B Partners405B PartnersHugging FaceKaggle
How-to Guides
OverviewDistillationFine TuningPromptingQuantizationResponsible Use Guide ResourcesValidationVision Capabilities
Deployment
OverviewAccelerator ManagementAutoscalingCost ComparisonCost ProjectionInfrastructure MigrationPrivate Cloud DeploymentProduction Deployment PipelinesRegulated Industry Self HostingSecurity in ProductionVersioning
Integration Guides
OverviewLangChainLlamaIndex
Llama Everywhere
OverviewRunning Meta Llama in the CloudRunning Meta Llama on LinuxRunning Meta Llama on MacRunning Meta Llama on Windows
Model Cards and Prompt Formats
OverviewLlama 4Llama 3.3Llama 3.2Llama 3.1Meta Llama 3Meta Llama 2Meta Code Llama 70BMeta Code LlamaLlama Guard 4Llama Guard 3Meta Llama Guard 2Meta Llama Guard 1Prompt GuardOther Models
Community Support
Community Support and ResourcesResources

Llama cookbooks

Agentic Workflows With Llama and LangGraphBuild Email Agent With Llama 3Build a GitHub Triaging Agent With Llama
Deploy Llama on vLLMExtend Llama to New LanguagesHow to Build Notebook Llama
Third Party Integrations With LlamaToolcalling With LlamaWhatsApp Llama 4 Bot

Llama videos

Auto-Analyze & Triage GitHub Issues With LlamaBuild Multi-Modal WhatsApp Chatbot With Llama 4Building With Llama 4Convert PDFs to Podcast With NotebookLlamaFine-Tuning Llama Models in the Real WorldGenerate Technical Blogs From Your Docs With Llama and RAGImplement Llama Protections to GenAI ApplicationsIntroducing Llama APIKnowledge Distillation With Llama 3.1 405BLlama 4 HerdLlama and Sourcegraph
Llama Fine-Tune Structured Image ExtractionLlama ToolboxLlamaCon 2025 KeynoteMore Ways to Run Llama 3Optimize Llama PromptsRunning Llama 3 on LinuxRunning Llama 3 on MacRunning Llama 3 on WindowsSynthetic Datasets for Fine-Tuning LlamaUnderstanding the Llama 3 Tokenizer

Llama case studies

AddAIAitemAitomatic SemiKongAitomaticAuthorshipBeneteBiofyBrain4DataCase Based ResearchCodeGPTContextual AICornerstoneDatabricks
Edgeless SystemsEnterprise Consulting PartnersExatiFirst AmericanInfo EdgeLeidosLitmosMajor Health SystemMajor Research UniversityNeoSpaceNorth Dakota LegislativeOxen AIOxide AI
ProaPwCRologyScribd IncSevilla Football ClubShopifySmartlySofyaStoqueTabnineTinfoilUpwork