Meta AI

New Capabilities

Responding to Gemma 4 with accelerated Llama 5 timeline (Q3 2026)

For two years, the most capable AI models lived behind paywalls and API meters. Google just made that harder to justify. Gemma 4, released April 2, is a family of four open-source models — ranging from 2 billion to 31 billion parameters — that handle text, images, video, and audio, run on consumer hardware, and ship under a fully permissive Apache 2.0 license with no usage restrictions.

Updated 14 minutes ago