{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreiaoiaxotiwf6x4of24djyvjnzi7nk74imyus3wn5rsi6q5676vrza",
    "uri": "at://did:plc:pgryn3ephfd2xgft23qokfzt/app.bsky.feed.post/3mkbci4rgfru2"
  },
  "path": "/t/building-local-my-2026-headless-ai-server-journey/175243#post_6",
  "publishedAt": "2026-04-24T18:25:49.000Z",
  "site": "https://discuss.huggingface.co",
  "textContent": "**\"I totally agree on the ‘pop up’ nature of these releases—it feels like we go months with small tweaks, and then a model like Gemma 4 just resets the baseline for what’s possible on consumer hardware.**\n\n**I’m actually restructuring my whole setup to give these new models more room to run. I’m moving to a three-node system:**\n\n  1. **AI Headless Server:** My gaming PC (7800 XT 16GB / 5600 CPU) dedicated 100% to the LLM weights. No display out, no background apps—just raw VRAM for the model.\n\n  2. **Middleware Server:** A Lenovo ThinkServer handling the ‘heavy lifting’ of the UI (Open WebUI), RAG/File processing (AnythingLLM), and the Cloudflare tunnel.\n\n  3. **Daily Driver:** My main PC just for the GUI.\n\n\n\n\n**My goal is to get the Gemma 4 26B (A4B) running at its full potential. By keeping the ‘Admin’ tasks on the ThinkServer, I’m hoping to keep that 26B model snappy (aiming for 20 t/s) while keeping the intelligence of a much larger model. It really feels like we’re finally reaching the point where local ‘mid-range’ hardware can compete with the big cloud models.\"**",
  "title": "Building Local: My 2026 Headless AI Server Journey"
}