Llama
Verified Oct 22, 2026 ActiveMeta's family of open-weight LLMs, downloadable under a community licence with a 700M-MAU carve-out.
Listed on EffectHub ·
Pricing
- Pricing model
- 100% Free
The weights cost nothing under the community licence. You pay for whatever runs them: your own GPUs, Meta's Model API (login or contact sales, no public price list seen) or a third-party host.
Use tool ↗About Llama
Llama is Meta's open-weight model line, from Llama 2 and the Llama 3.1 to 3.3 text models to the natively multimodal Llama 4 mixture-of-experts releases, with Llama Guard and Prompt Guard as safety add-ons. Weights are free after accepting the licence, via Meta's partners, Hugging Face or Kaggle, and the same checkpoints run on llama.cpp, vLLM and most inference stacks; Meta also sells hosted access through its Model API. The Llama 4 Community License permits commercial use below 700 million monthly active users, requires a 'Built with Llama' notice and a 'Llama' prefix on derivative model names, and names Meta Platforms, Inc. (Meta Platforms Ireland for the EEA) as licensor. The web home has moved again: llama.com and llama.meta.com now forward to Meta's developer site, where Llama sits beside the newer Muse model family.
- Platforms
- Web · API
- API
- Yes
- Last verified
- Oct 22, 2026
Verified sources
- www.llama.com and its /docs, /llama-downloads and /license paths return 301 to developer.meta.com/ai and on to dev.meta.ai (15 October 2026)llama.com ↗
- Llama docs on Meta's developer site list Llama 4 (multimodal MoE), Llama 3.1 to 3.3, Llama 2, Llama Guard 4 and Prompt Guard, with weights via partners, Hugging Face and Kaggle and hosted access through the Meta Model API; the Muse family is listed alongside (15 October 2026)dev.meta.ai ↗
- Llama 4 Community License (effective 5 April 2025) names Meta Platforms, Inc. or Meta Platforms Ireland Limited as licensor, requires a separate licence above 700M MAU, 'Built with Llama' attribution and a 'Llama' prefix on derivative models (15 October 2026)dev.meta.ai ↗
Is Llama legit?
We don’t label tools safe or unsafe. Below are the records we checked ourselves, each with the date we read it.
- Domain registered
- Nov 1, 1994 (31 years ago)Registry record (RDAP), registrar RegistrarSEC LLC, looked up Sep 6, 2026
- TLS certificate
- Issued by DigiCert Inc, valid until Oct 30, 2026Seen in our own connection to the site on Oct 22, 2026
- Last checked
- Oct 22, 2026The site answered
- On the market since
- 2023The vendor's own material, read by us
- Operated by
- Meta Platforms, Inc., United StatesOur research, card verified Oct 15, 2026
Pros & cons of Llama
Still the default choice when you need weights you can hold, fine-tune and run offline, with an ecosystem of quantisations and inference stacks built around it. The licence is not OSI-open, and Meta's web presence keeps moving: llama.com now forwards wholesale to the developer site, where Llama is listed next to the newer Muse family, so links rot faster than the models do and the roadmap looks less Llama-first than it did.
Pros
- Weights are downloadable, not just API-accessible
- Broad tooling support: llama.cpp, vLLM, Hugging Face and most inference runtimes
- Multiple sizes so you can trade quality against a single GPU
- Commercial use permitted for almost everyone under the MAU threshold
Cons
- Community licence is not open source; attribution, the Llama name prefix and the MAU cap all bind
- Meta keeps moving the site: llama.com and llama.meta.com now forward to dev.meta.ai, where Llama shares the menu with the Muse models
- Running it is your cost and your operational problem
- Meta's Model API shows no public price; it is login or contact sales
Best for
- Self-hosting a capable model on your own hardware
- Fine-tuning on private data that cannot leave your network
- Teams wanting an exit from per-token API pricing
Not for
- Products above 700M monthly active users without a separate Meta licence
- Anyone who needs a strictly OSI-approved licence
Watch out: llama.com, llama.meta.com and the old /docs paths all redirect to dev.meta.ai now; bookmark dev.meta.ai/llama for the docs and licences, and read the Model API terms separately because no price is published.
Alternatives to Llama
Llama head-to-head
Side-by-side fact sheets against the competitors our research or their makers name — pricing, platforms, API and both verdicts on one page.