Meta's largest open model: closed-frontier quality but datacenter requirements. In practice it needs high-VRAM multi-GPU or the cloud.
Meta
Meta Platforms (Menlo Park, California, USA): founded in 2004 as Facebook, it publishes the Llama family — the world's most widely used open-weight LLM line, from 1B to 405B parameters, under a commercial-use license. The Llama models in the catalog are the child resources.
Features
- Llama 3.1/3.2/4: from 1B to 405B parameters
- Meta Llama license for commercial use
- Native support on CPU, CUDA GPU and Apple Silicon
Technical specifications
- Headquarters
- Menlo Park, California (USA)
- Founded
- 2004
- Employees
- ~72,000
- Revenue
- 201 bn USD (2025)
- Ownership
- Public company (Nasdaq: META)
Resources by Meta
Frontier-class open model with 131K-token context. Needs 48-80GB VRAM, a Mac with unified memory, or a multi-GPU setup.
The de-facto local workhorse: solid quality, 131K-token context and a huge GGUF quantization ecosystem. The standard choice for chat and assistants.
Small, fast model with up to 131K-token context. Ideal entry point for chat and summarization on modest hardware, great on phones and mini-PCs.
Il modello MoE grande della famiglia Llama 4 di Meta: 400B di parametri totali con 17B attivi (128 esperti). Contesto 1M di token, nativamente multimodale (testo+immagini), licenza comunitaria Meta. Richiede multi-GPU o macchine con memoria unificata 128GB+ anche quantizzato.
The compact MoE of the Llama 4 family: 17B active parameters out of 109B total (16 experts). Meta's community license allows free use under 700 million monthly users. One of the best entry points into the Llama 4 family.
Generates up to 30 seconds of music or sound effects from a text description. The open reference for music models: light and versatile.
Tags
Discover Meta
Visit the official site to download, learn more and start using Meta.

