richarderkhov/toheartzhang_-_jiuzhang3.0-synthesis-7b-gguf overview

{
  "metadata": {},
  "card_data": {
    "frontmatter": {},
    "hero_image_url": "",
    "summary": "Quantization made by Richard Erkhov. Github Discord Request more models JiuZhang3.0-Synthesis-7B - GGUF | Name | Quant method | Size | | ---- | ---- | ---- | | JiuZhang3.0-Synthesis-7B.Q2_K.gguf | Q2_K | 2.53GB | | JiuZhang3.0-Synthesis-7B.IQ3_XS.gguf | IQ3_XS | 2.79GB | | JiuZhang3.0-Synthesis-7B.IQ3_S.gguf | IQ3_S | 2.92GB | | JiuZhang3.0-Synthesis-7B.Q3_K_S.gguf | Q3_K_S | 2.92GB | | JiuZhang3.0-Synthesis-7B.IQ3_M.gguf | IQ3_M | 3.06GB | | JiuZhang3.0-Synthesis-7B.Q3_K.gguf | Q3_K | 3.22GB | | JiuZhang3.0-Synthesis-7B.Q3_K_M.gguf | Q3_K_M | 3.22GB | | JiuZhang3.0-Synthesis-7B.Q3_K_L.gguf | Q3_K_L | 3.49GB | | JiuZhang3.0-Synthesis-7B.IQ4_XS.gguf | IQ4_XS | 3.56GB | | JiuZhang3.0-Synthesis-7B.Q4_0.gguf | Q4_0 | 3.73GB | | JiuZhang3.0-Synthesis-7B.IQ4_NL.gguf | IQ4_NL | 3.74GB | | JiuZhang3.0-Synthesis-7B.Q4_K_S.gguf | Q4_K_S | 3.75GB | | JiuZhang3.0-Synthesis-7B.Q4_K.gguf | Q4_K | 3.93GB | | JiuZhang3.0-Synthesis-7B.Q4_K_M.gguf | Q4_K_M | 3.93GB | | JiuZhang3.0-Synthesis-7B.Q4_1.gguf | Q4_1 | 4.1GB | | JiuZhang3.0-Synthesis-7B.Q5_0.gguf | Q5_0 | 4.48GB | | JiuZhang3.0-Synthesis-7B.Q5_K_S.gguf | Q5_K_S | 4.48GB | | JiuZhang3.0-Synthesis-7B.Q5_K.gguf | Q5_K | 4.59GB | | JiuZhang3.0-Synthesis-7B.Q5_K_M.gguf | Q5_K_M | 4.59GB | | JiuZhang3.0-Synthesis-7B.Q5_1.gguf | Q5_1 | 4.86GB | | JiuZhang3.0-Synthesis-7B.Q6_K.gguf | Q6_K | 5.28GB | | JiuZhang3.0-Synthesis-7B.Q8_0.gguf | Q8_0 | 6.84GB | Original model description: --- language: ---  JiuZhang3.0: Efficiently Improving Mathematical Reasoning by Training Small Data Synthesis Models   [Paper] • [GitHub] • [Models] • [Data]",
    "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\nJiuZhang3.0-Synthesis-7B - GGUF\n- Model creator: https://huggingface.co/ToheartZhang/\n- Original model: https://huggingface.co/ToheartZhang/JiuZhang3.0-Synthesis-7B/\n\n\n| Name | Quant method | Size |\n| ---- | ---- | ---- |\n| [JiuZhang3.0-Synthesis-7B.Q2_K.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.Q2_K.gguf) | Q2_K | 2.53GB |\n| [JiuZhang3.0-Synthesis-7B.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.IQ3_XS.gguf) | IQ3_XS | 2.79GB |\n| [JiuZhang3.0-Synthesis-7B.IQ3_S.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.IQ3_S.gguf) | IQ3_S | 2.92GB |\n| [JiuZhang3.0-Synthesis-7B.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.Q3_K_S.gguf) | Q3_K_S | 2.92GB |\n| [JiuZhang3.0-Synthesis-7B.IQ3_M.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.IQ3_M.gguf) | IQ3_M | 3.06GB |\n| [JiuZhang3.0-Synthesis-7B.Q3_K.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.Q3_K.gguf) | Q3_K | 3.22GB |\n| [JiuZhang3.0-Synthesis-7B.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.Q3_K_M.gguf) | Q3_K_M | 3.22GB |\n| [JiuZhang3.0-Synthesis-7B.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.Q3_K_L.gguf) | Q3_K_L | 3.49GB |\n| [JiuZhang3.0-Synthesis-7B.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.IQ4_XS.gguf) | IQ4_XS | 3.56GB |\n| [JiuZhang3.0-Synthesis-7B.Q4_0.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.Q4_0.gguf) | Q4_0 | 3.73GB |\n| [JiuZhang3.0-Synthesis-7B.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.IQ4_NL.gguf) | IQ4_NL | 3.74GB |\n| [JiuZhang3.0-Synthesis-7B.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.Q4_K_S.gguf) | Q4_K_S | 3.75GB |\n| [JiuZhang3.0-Synthesis-7B.Q4_K.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.Q4_K.gguf) | Q4_K | 3.93GB |\n| [JiuZhang3.0-Synthesis-7B.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.Q4_K_M.gguf) | Q4_K_M | 3.93GB |\n| [JiuZhang3.0-Synthesis-7B.Q4_1.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.Q4_1.gguf) | Q4_1 | 4.1GB |\n| [JiuZhang3.0-Synthesis-7B.Q5_0.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.Q5_0.gguf) | Q5_0 | 4.48GB |\n| [JiuZhang3.0-Synthesis-7B.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.Q5_K_S.gguf) | Q5_K_S | 4.48GB |\n| [JiuZhang3.0-Synthesis-7B.Q5_K.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.Q5_K.gguf) | Q5_K | 4.59GB |\n| [JiuZhang3.0-Synthesis-7B.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.Q5_K_M.gguf) | Q5_K_M | 4.59GB |\n| [JiuZhang3.0-Synthesis-7B.Q5_1.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.Q5_1.gguf) | Q5_1 | 4.86GB |\n| [JiuZhang3.0-Synthesis-7B.Q6_K.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.Q6_K.gguf) | Q6_K | 5.28GB |\n| [JiuZhang3.0-Synthesis-7B.Q8_0.gguf](https://huggingface.co/RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf/blob/main/JiuZhang3.0-Synthesis-7B.Q8_0.gguf) | Q8_0 | 6.84GB |\n\n\n\n\nOriginal model description:\n---\nlanguage:\n- en\n---\n<h1 align=\"center\">\nJiuZhang3.0: Efficiently Improving Mathematical\nReasoning by Training Small Data Synthesis Models\n</h1>\n<p align=\"center\">\n  <a href=\"https://arxiv.org/abs/2405.14365\"><b>[Paper]</b></a> •\n  <a href=\"https://github.com/RUCAIBox/JiuZhang3.0\"><b>[GitHub]</b></a> •\n  <a href=\"https://huggingface.co/collections/ToheartZhang/jiuzhang30-66508be8be5a61de47101655#/\"><b>[Models]</b></a> •\n  <a href=\"https://huggingface.co/collections/ToheartZhang/jiuzhang30-corpus-665092209525389ad7a2289a\"><b>[Data]</b></a>\n</p>\n\n## Introduction\nJiuZhang3.0 is a series of fine-tuned models for math reasoning continually pre-trained on corpus synthesized by our carefully trained small LLM.\n\n## Experimental Results\nFor more evaluation results, please refer to the [Paper](https://arxiv.org/abs/2405.14365)\n\n| Models                   | GSM8k | MATH | SVAMP | ASDiv | MAWPS | CARP | Avg.  |\n|--------------------------|-------|------|-------|-------|-------|------|-------|\n| GPT-4                    | 92.2  | 65.4 | 92.9  | 94.3  | 96.6  | 53.6 | 82.5  |\n|**20B+ Models**||\n| Llemma-34B               | 60.2  | 24.6 | 68.0  | 75.6  | 89.8  | 36.5 | 59.1  |\n| Intern-Math-20B          | 64.9  | 27.4 | 74.9  | 79.6  | 94.4  | 42.3 | 63.9  |\n| ChatGLM-Math-32B         | 82.6  | 40.6 | -     | -     | -     | -    | -     |\n| MAmmoTH2-8x7B-Plus       | _86.4_| 47.0 | _90.0_| _92.2_| **97.0** | 45.8 | _76.4_ |\n| [JiuZhang3.0-8x7B](https://huggingface.co/ToheartZhang/JiuZhang3.0-8x7B)   | **89.8** | **53.8** | **90.2** | **93.1** | _96.7_ | 52.3 | **79.3** |\n|**7-8B Models**||\n| Mistral-7B-MMIQC         | 75.0  | 34.2 | 73.5  | 82.1  | 90.1  | 36.5 | 65.2  |\n| MetaMath-Mistral-7B      | 77.8  | 29.6 | 79.6  | 81.2  | 93.7  | 30.5 | 65.4  |\n| Abel-7B-002              | 80.4  | 29.6 | 78.8  | 82.7  | 93.5  | 33.2 | 66.4  |\n| WizardMath-7B-1.1        | 82.2  | 32.8 | 80.7  | 84.2  | 93.8  | 31.9 | 67.6  |\n| Math-Shepherd-Mistral-7B | 84.3  | 34.4 | 82.9  | 82.8  | 92.5  | 32.9 | 68.3  |\n| KPMath-DSMath-7B         | 83.9  | 48.8 | 81.5  | 88.9  | 94.8  | -    | -     |\n| MAmmoTH2-7B-Plus         | 84.2  | 46.2 | _90.3_| 90.3  | _97.1_| 44.3 | 75.2  |\n| MAmmoTH2-8B-Plus         | 84.4  | 41.2 | 89.9  | 89.9  | _97.1_| 44.8 | 74.6  |\n| DeepSeekMath-7B-Instruct | 82.3  | 45.8 | 83.7  | 90.1  | 95.7  | 45.8 | 73.9  |\n| DeepSeekMath-7B-RL       | 88.2  | 50.2 | 87.3  | 91.8  | 95.5  | **51.6** | 77.4  |\n| [JiuZhang3.0-7B](https://huggingface.co/ToheartZhang/JiuZhang3.0-7B)     | **88.6** | **52.8** | **90.4** | **92.6** | **97.3** | _51.0_ | **78.8** |\n| [JiuZhang3.0-8B](https://huggingface.co/ToheartZhang/JiuZhang3.0-8B)    | **88.6** | _51.0_ | 89.4  | **92.6** | _97.1_ | 50.9 | _78.3_ |\n\n## Data Format\nThis is a fine-tuned model for synthesizing math-related question-solution pairs.\nPlease use the given [prompts](https://github.com/RUCAIBox/JiuZhang3.0/tree/main/prompts/cot).\n\n## Syntheized Data Type\n* Grade school\n* Middel school\n* High school\n* College\n* AMC 8\n* AMC 10\n* AMC 12\n* AIME\n\n\n## Citation\nIf you find this repository helpful, please consider citing our paper:\n\n```\n@article{zhou2024jiuzhang30,\n      title={JiuZhang3.0: Efficiently Improving Mathematical Reasoning by Training Small Data Synthesis Models}, \n      author={Kun Zhou and Beichen Zhang and Jiapeng Wang and Zhipeng Chen and Wayne Xin Zhao and Jing Sha and Zhichao Sheng and Shijin Wang and Ji-Rong Wen},\n      year={2024},\n}\n```\n\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "arxiv:2405.14365",
    "endpoints_compatible",
    "region:us",
    "conversational"
  ],
  "likes": 0,
  "downloads": 508,
  "gated": false,
  "private": false,
  "last_modified": "2024-09-19T21:53:19.000Z",
  "created_at": "2024-09-19T17:33:35.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "66ec606f4cec785689bf2560",
  "id": "RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf",
  "modelId": "RichardErkhov/ToheartZhang_-_JiuZhang3.0-Synthesis-7B-gguf",
  "sha": "775db066d18d545dc98ee0314ab8e4c0f3f361b7",
  "createdAt": "2024-09-19T17:33:35.000Z",
  "lastModified": "2024-09-19T21:53:19.000Z",
  "author": "RichardErkhov",
  "downloads": 508,
  "likes": 0,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 24
}