What This Documents
Llama 4 is Meta's fourth-generation Llama model family, released with open weights and documented through Meta's own technical materials, including variants publicly referred to as Scout and Maverick.
What's Publicly Known
Meta has historically emphasized open-weight releases for the Llama family, and Llama 4 continued this pattern, making the model weights publicly downloadable alongside accompanying technical documentation.
Public and developer reception to the initial Llama 4 release was reportedly mixed on some fronts, which has been cited in industry reporting as a factor in Meta's subsequent reorganization of its AI research efforts; see our Meta news page for more on this reported context.
Verify Current Details
Treat the summary above as a general orientation rather than a substitute for the primary source — for exact technical details, benchmark figures, and methodology, read Meta's own published documentation directly.
Related Pages
Frequently Asked
Is Llama 4 open-weight?
Yes, consistent with Meta's historical approach to the Llama family.
What are Scout and Maverick?
Publicly referenced variant names within the Llama 4 release; check Meta's official documentation for the current specific lineup and differences between them.
How was Llama 4 received?
Industry reporting has described a mixed developer response on some fronts; see our Meta news page for more current context on this.
Where can I read Meta's official technical materials?
Check Meta AI's official site for current published documentation on the Llama 4 family.