richarderkhov/infinirc_-_infinirc-llama3-8b-2g-release-v1.0-gguf overview

{
  "metadata": {},
  "card_data": {
    "frontmatter": {},
    "hero_image_url": "",
    "summary": "",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "Quantization made by Richard Erkhov.\n\n[Github](https://github.com/RichardErkhov)\n\n[Discord](https://discord.gg/pvy7H8DZMG)\n\n[Request more models](https://github.com/RichardErkhov/quant_request)\n\n\nInfinirc-Llama3-8B-2G-Release-v1.0 - GGUF\n- Model creator: https://huggingface.co/Infinirc/\n- Original model: https://huggingface.co/Infinirc/Infinirc-Llama3-8B-2G-Release-v1.0/\n\n\n| Name | Quant method | Size |\n| ---- | ---- | ---- |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.Q2_K.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.Q2_K.gguf) | Q2_K | 2.96GB |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.IQ3_XS.gguf) | IQ3_XS | 3.28GB |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.IQ3_S.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.IQ3_S.gguf) | IQ3_S | 3.43GB |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.Q3_K_S.gguf) | Q3_K_S | 3.41GB |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.IQ3_M.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.IQ3_M.gguf) | IQ3_M | 3.52GB |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.Q3_K.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.Q3_K.gguf) | Q3_K | 3.74GB |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.Q3_K_M.gguf) | Q3_K_M | 3.74GB |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.Q3_K_L.gguf) | Q3_K_L | 4.03GB |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.IQ4_XS.gguf) | IQ4_XS | 4.18GB |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.Q4_0.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.Q4_0.gguf) | Q4_0 | 4.34GB |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.IQ4_NL.gguf) | IQ4_NL | 4.38GB |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.Q4_K_S.gguf) | Q4_K_S | 4.37GB |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.Q4_K.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.Q4_K.gguf) | Q4_K | 4.58GB |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.Q4_K_M.gguf) | Q4_K_M | 4.58GB |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.Q4_1.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.Q4_1.gguf) | Q4_1 | 4.78GB |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.Q5_0.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.Q5_0.gguf) | Q5_0 | 5.21GB |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.Q5_K_S.gguf) | Q5_K_S | 5.21GB |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.Q5_K.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.Q5_K.gguf) | Q5_K | 5.34GB |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.Q5_K_M.gguf) | Q5_K_M | 5.34GB |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.Q5_1.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.Q5_1.gguf) | Q5_1 | 5.65GB |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.Q6_K.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.Q6_K.gguf) | Q6_K | 6.14GB |\n| [Infinirc-Llama3-8B-2G-Release-v1.0.Q8_0.gguf](https://huggingface.co/RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf/blob/main/Infinirc-Llama3-8B-2G-Release-v1.0.Q8_0.gguf) | Q8_0 | 7.95GB |\n\n\n\n\nOriginal model description:\n---\nlicense: llama3\nlanguage:\n- zh\n- en\nlibrary_name: transformers\ntags:\n- zhtw\n---\n\n\n\n\n\n# Llama3 8B (台灣文化版) \n\n## 模型詳情\n**開發者**：陳昭儒[Infinirc.com](https://infinirc.com)  \n**模型版本**：1.0  \n**模型類型**：自然語言處理  \n**訓練數據源**：采用與台灣文化相關的資料集，包括台灣新聞、文學作品、網路文章等。\n\n## 目的和用途\n這款Llama3 8B模型是專門為了更好地理解和生成與台灣文化相關的文本而設計和微調的。目標是提供一個能夠捕捉台灣特有文化元素和語言習慣的強大語言模型，適用於文本生成、自動回答等多種應用。\n\n## 模型架構\n**基礎模型**：Llama3 8B  \n**調整策略**：對模型進行微調，使用與台灣文化相關的具體資料集進行微調，以增強模型對於本地化內容的理解和生成能力。\n\n## 性能指標\n該模型在一系列NLP基準測試中展示了優異的性能，特別是在文本生成和語意理解方面具有較高的準確度。  \n**具體性能數據**：(詳述BLEU分數、ROUGE分數等性能數據)\n\n## 使用和限制\n**使用建議**：推薦於需要深度理解或生成與台灣文化相關內容的應用場景中使用。  \n**限制**：雖然模型針對台灣文化進行了特定優化，但在處理非目標區域的文化內容時可能效果不佳。\n\n## 風險與倫理考量\n使用本模型時應注意確保生成的內容不包含歧視性或有害信息。模型的開發和使用應遵循倫理準則和社會責任。\n\n## 聯絡方式\n如有任何問題或需要進一步的信息，請透過下方聯絡方式與我們團隊聯繫：  \nEmail: [ricky@infinirc.com](mailto:ricky@infinirc.com)  \n網站: [https://infinirc.com](https://infinirc.com)\n\n\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "endpoints_compatible",
    "region:us",
    "conversational"
  ],
  "likes": 0,
  "downloads": 179,
  "gated": false,
  "private": false,
  "last_modified": "2024-08-22T09:57:44.000Z",
  "created_at": "2024-08-22T08:03:22.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "66c6f0cab7da6985260635a2",
  "id": "RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf",
  "modelId": "RichardErkhov/Infinirc_-_Infinirc-Llama3-8B-2G-Release-v1.0-gguf",
  "sha": "91321d911f82e0a8da028d7e0db96470aeed1f21",
  "createdAt": "2024-08-22T08:03:22.000Z",
  "lastModified": "2024-08-22T09:57:44.000Z",
  "author": "RichardErkhov",
  "downloads": 179,
  "likes": 0,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 24
}