Meta

Owner of the open-source Llama family of models, with a focus on accessibility and transparency in AI development.

6 Featured Models

6 Featured Text Models

Compare Models

Key Highlights

Originally Facebook AI Research (FAIR), founded in 2013, led by Yann LeCun.

Focuses on fundamental AI research (NLP, CV, RL) and open science.

Developed and open-sourced the popular PyTorch deep learning framework.

Champions open-source LLMs with the Llama family (Llama, Llama 2, Llama 3, Llama 4).

Llama 4 (Scout, Maverick, Behemoth) features native multimodality and Mixture-of-Experts (MoE).

Llama 4 Scout has a 10 million token context window.

Developed the Meta AI assistant integrated into Facebook, Instagram, WhatsApp, Messenger, Ray-Ban Meta glasses, and Quest VR headsets.

Research includes protein folding (predicting 600M structures), conversational AI, and unsupervised machine translation.

Developing custom AI hardware (MTIA v1 accelerator).

Contributes to AI safety tools like Llama Guard.

Llama 4 Maverick

Apr 2025

Llama 4 Maverick is Meta's multimodal expert model with 17B active parameters and 128 experts (400B total parameters). It outperforms GPT-4o and Gemini 2.0 Flash across various benchmarks, achieving an ELO of 1417 on LMArena. Designed for sophisticated AI applications with excellent image understanding and creative writing.

conversationreasoningcode-generationanalysisplanning

Llama 4 Scout

Apr 2025

Llama 4 Scout is Meta's compact yet powerful multimodal model with 17B active parameters and 16 experts (109B total parameters). It fits on a single H100 GPU with Int4 quantization and offers an industry-leading 10M token context window, outperforming Gemma 3, Gemini 2.0 Flash-Lite, and Mistral 3.1 across various benchmarks.

conversationreasoningcode-generationanalysis

Llama 3.1 70B

Jul 2024

Llama 3.1 70B offers a dramatically expanded context window and improved performance on mathematical reasoning and general knowledge tasks.

conversationreasoningcode-generation

Llama 3.1 405B

Jul 2024

Llama 3.1 405B is Meta's most powerful open-source model, outperforming even proprietary models on various benchmarks.

conversationreasoningcode-generationanalysis

Llama 3 70B

Apr 2024

Llama 3 70B is a large language model from Meta with strong performance and efficiency for real-time interactions.

conversationreasoningcode-generation

Llama 4 Behemoth

Invalid Date

Llama 4 Behemoth is Meta's most powerful model yet with 288B active parameters and 16 experts (nearly 2T total parameters), outperforming GPT-4.5, Claude Sonnet 3.7, and Gemini 2.0 Pro on several STEM benchmarks.

conversationreasoningcode-generationanalysis