{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreidimzs5tyngaaom4uud57dmvzwciisimaa6dtpeyxw7ko2iksbz3u",
    "uri": "at://did:plc:pgryn3ephfd2xgft23qokfzt/app.bsky.feed.post/3mnnmxq2u4pd2"
  },
  "path": "/t/fine-tuning-an-slm-for-a-low-resource-language/176467#post_8",
  "publishedAt": "2026-06-06T21:14:25.000Z",
  "site": "https://discuss.huggingface.co",
  "tags": [
    "Gemma is basically like Gemini’s younger sibling",
    "Open Persian LLM Leaderboard - a Hugging Face Space by opll-org",
    "Open Persian LLM Leaderboard - a Hugging Face Space by PartAI",
    "MIZAN: A Persian LLM Leaderboard - a Hugging Face Space by MCINext",
    "GitHub - ParsBench/ParsBench: ParsBench provides toolkits for benchmarking LLMs based on the Persian language tasks. · GitHub",
    "ParsBench (ParsBench)",
    "[2502.11571] FaMTEB: Massive Text Embedding Benchmark in Persian Language",
    "amirakhlaghiqqq/PersianPhi · Hugging Face"
  ],
  "textContent": "Oh. I don’t know Persian myself, but Gemma is basically like Gemini’s younger sibling, so my guess is that it may be fairly strong for Persian too. Also, leaderboards can make the search much easier:\n\n  * Open Persian LLM Leaderboard / OPLL\nOpen Persian LLM Leaderboard - a Hugging Face Space by opll-org\n\n  * Another Open Persian LLM Leaderboard mirror / variant\nOpen Persian LLM Leaderboard - a Hugging Face Space by PartAI\n\n  * MIZAN: Persian LLM Leaderboard\nMIZAN: A Persian LLM Leaderboard - a Hugging Face Space by MCINext\n\n  * ParsBench\nGitHub - ParsBench/ParsBench: ParsBench provides toolkits for benchmarking LLMs based on the Persian language tasks. · GitHub\nParsBench (ParsBench)\n\n  * FaMTEB, if embeddings / RAG are relevant\n[2502.11571] FaMTEB: Massive Text Embedding Benchmark in Persian Language\n\n  * PersianPhi may also be worth checking as a Persian-adapted compact model\namirakhlaghiqqq/PersianPhi · Hugging Face\n\n\n\n\nSo, if me, I’d probably use those leaderboards to make a shortlist, then compare tokenization and a small private Persian eval before choosing the base model.",
  "title": "Fine-Tuning an SLM for a Low-Resource Language"
}