{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreicaktfag2kgndj6kujexx77spluemhft64z3w322n2qfowawjuule",
    "uri": "at://did:plc:pgryn3ephfd2xgft23qokfzt/app.bsky.feed.post/3mlw5o7lenhh2"
  },
  "path": "/t/need-english-only-or-minimal-multilingual-2b-4b-llm-for-agentic-ai-on-gtx-1660-super-6gb-vram-quantization-friendly/176044#post_4",
  "publishedAt": "2026-05-15T18:34:33.000Z",
  "site": "https://discuss.huggingface.co",
  "textContent": "Glad to hear that, azhak1!\n\nLooking forward to your benchmarks. Pay special attention to the context window stability—on 6GB cards, that’s where the real ‘agentic’ battle happens. If you run into CUDA OOM errors, let us know, we have some custom cleanup routines that might help.\n\nStay tuned.",
  "title": "Need English-only (or minimal multilingual) 2B-4B LLM for Agentic AI on GTX 1660 Super (6GB VRAM) – quantization friendly"
}