{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreidjqnsihxsnax7j5xyzar72qpec5k3lghxlh6ce34765dcmdd4ivq",
    "uri": "at://did:plc:avkh7zze5iapdkk6naaunrjn/app.bsky.feed.post/3mngd5hlko2d2"
  },
  "path": "/260603/p58#a260603p58",
  "publishedAt": "2026-06-03T22:25:03.000Z",
  "site": "https://www.techmeme.com",
  "tags": [
    "VentureBeat",
    "Google introduces Gemma 4 12B, a unified, encoder-free open multimodal model that can run locally on devices with 16GB of VRAM or unified memory"
  ],
  "textContent": "Carl Franzen / VentureBeat:\n**Google introduces Gemma 4 12B, a unified, encoder-free open multimodal model that can run locally on devices with 16GB of VRAM or unified memory** — While many AI open source model providers are pursuing larger and more powerful models, Google is still giving attention to the smaller, more local side of the market.",
  "title": "Google introduces Gemma 4 12B, a unified, encoder-free open multimodal model that can run locally on devices with 16GB of VRAM or unified memory (Carl Franzen/VentureBeat)"
}