prithivmlmods/dolphin-v2-f32-gguf overview

{
  "metadata": {},
  "card_data": {
    "license": "apache-2.0",
    "language": [
      "en",
      "zh"
    ],
    "base_model": [
      "ByteDance/Dolphin-v2"
    ],
    "pipeline_tag": "image-text-to-text",
    "library_name": "transformers",
    "tags": [
      "text-generation-inference",
      "document-parsing",
      "document-understanding",
      "document-intelligence",
      "ocr",
      "layout-analysis",
      "table-extraction",
      "formula-recognition",
      "code-extraction",
      "vision-language-model",
      "multimodal"
    ],
    "frontmatter": {
      "license": "apache-2.0",
      "language": [
        "en",
        "zh"
      ],
      "base_model": [
        "ByteDance/Dolphin-v2"
      ],
      "pipeline_tag": "image-text-to-text",
      "library_name": "transformers",
      "tags": [
        "text-generation-inference",
        "document-parsing",
        "document-understanding",
        "document-intelligence",
        "ocr",
        "layout-analysis",
        "table-extraction",
        "formula-recognition",
        "code-extraction",
        "vision-language-model",
        "multimodal"
      ]
    },
    "hero_image_url": "https://www.nethype.de/huggingface_embed/quantpplgraph.png",
    "summary": "> ByteDance Dolphin-v2 is a 3B-parameter vision-language model built on Qwen2.5-VL-3B with Native Resolution Vision Transformer (NaViT) encoder and autoregressive decoder, designed as a universal document parsing solution via a document-type-aware two-stage architecture that classifies digital-born vs. photographed documents before applying hybrid strategies—element-wise parallel parsing for clean PDFs and holistic parsing for distorted scans. It supports 21 element categories (headings sec_0-5, paragraphs, formulas in LaTeX, HTML tables, indented code blocks, figures, lists, etc.) with absolute pixel coordinates for precise localization, achieving state-of-the-art OmniDocBench v1.5 scores of 89.45 overall (+14.78 over original Dolphin), 0.054 edit distance for text/reading order, 86.72% CDM for formulas, and 87.02/90.48 TEDS/TEDS-S for tables at 0.1729 FPS on 8-12GB VRAM GPUs. Specialized modules (P_formula, P_code, P_table, P_paragraph) enable structured JSON/Markdown/HTML outputs for privacy-focused local inference in healthcare/legal/finance, outperforming general VLMs in speed (2x faster) and accuracy across distortions, skews, and perspectives.",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "---\nlicense: apache-2.0\nlanguage:\n- en\n- zh\nbase_model:\n- ByteDance/Dolphin-v2\npipeline_tag: image-text-to-text\nlibrary_name: transformers\ntags:\n- text-generation-inference\n- document-parsing\n- document-understanding\n- document-intelligence\n- ocr\n- layout-analysis\n- table-extraction\n- formula-recognition\n- code-extraction\n- vision-language-model\n- multimodal\n---\n\n# **Dolphin-v2-f32-GGUF**\n\n> ByteDance Dolphin-v2 is a 3B-parameter vision-language model built on Qwen2.5-VL-3B with Native Resolution Vision Transformer (NaViT) encoder and autoregressive decoder, designed as a universal document parsing solution via a document-type-aware two-stage architecture that classifies digital-born vs. photographed documents before applying hybrid strategies—element-wise parallel parsing for clean PDFs and holistic parsing for distorted scans. It supports 21 element categories (headings sec_0-5, paragraphs, formulas in LaTeX, HTML tables, indented code blocks, figures, lists, etc.) with absolute pixel coordinates for precise localization, achieving state-of-the-art OmniDocBench v1.5 scores of 89.45 overall (+14.78 over original Dolphin), 0.054 edit distance for text/reading order, 86.72% CDM for formulas, and 87.02/90.48 TEDS/TEDS-S for tables at 0.1729 FPS on 8-12GB VRAM GPUs. Specialized modules (P_formula, P_code, P_table, P_paragraph) enable structured JSON/Markdown/HTML outputs for privacy-focused local inference in healthcare/legal/finance, outperforming general VLMs in speed (2x faster) and accuracy across distortions, skews, and perspectives.\n\n## Dolphin-v2 [GGUF]\n\n| File Name | Quant Type | File Size | File Link |\n| - | - | - | - |\n| Dolphin-v2.BF16.gguf | BF16 | 6.18 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.BF16.gguf) |\n| Dolphin-v2.F32.gguf | F32 | 12.3 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.F32.gguf) |\n| Dolphin-v2.IQ4_XS.gguf | IQ4_XS | 1.75 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.IQ4_XS.gguf) |\n| Dolphin-v2.Q2_K.gguf | Q2_K | 1.27 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.Q2_K.gguf) |\n| Dolphin-v2.Q3_K_L.gguf | Q3_K_L | 1.71 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.Q3_K_L.gguf) |\n| Dolphin-v2.Q3_K_M.gguf | Q3_K_M | 1.59 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.Q3_K_M.gguf) |\n| Dolphin-v2.Q3_K_S.gguf | Q3_K_S | 1.45 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.Q3_K_S.gguf) |\n| Dolphin-v2.Q4_K_M.gguf | Q4_K_M | 1.93 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.Q4_K_M.gguf) |\n| Dolphin-v2.Q4_K_S.gguf | Q4_K_S | 1.83 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.Q4_K_S.gguf) |\n| Dolphin-v2.Q5_K_M.gguf | Q5_K_M | 2.22 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.Q5_K_M.gguf) |\n| Dolphin-v2.Q5_K_S.gguf | Q5_K_S | 2.17 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.Q5_K_S.gguf) |\n| Dolphin-v2.Q6_K.gguf | Q6_K | 2.54 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.Q6_K.gguf) |\n| Dolphin-v2.Q8_0.gguf | Q8_0 | 3.29 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.Q8_0.gguf) |\n| Dolphin-v2.f16.gguf | F16 | 6.18 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.f16.gguf) |\n| Dolphin-v2.i1-IQ1_M.gguf | i1-IQ1_M | 850 MB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-IQ1_M.gguf) |\n| Dolphin-v2.i1-IQ1_S.gguf | i1-IQ1_S | 791 MB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-IQ1_S.gguf) |\n| Dolphin-v2.i1-IQ2_M.gguf | i1-IQ2_M | 1.14 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-IQ2_M.gguf) |\n| Dolphin-v2.i1-IQ2_S.gguf | i1-IQ2_S | 1.06 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-IQ2_S.gguf) |\n| Dolphin-v2.i1-IQ2_XS.gguf | i1-IQ2_XS | 1.03 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-IQ2_XS.gguf) |\n| Dolphin-v2.i1-IQ2_XXS.gguf | i1-IQ2_XXS | 948 MB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-IQ2_XXS.gguf) |\n| Dolphin-v2.i1-IQ3_M.gguf | i1-IQ3_M | 1.49 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-IQ3_M.gguf) |\n| Dolphin-v2.i1-IQ3_S.gguf | i1-IQ3_S | 1.46 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-IQ3_S.gguf) |\n| Dolphin-v2.i1-IQ3_XS.gguf | i1-IQ3_XS | 1.39 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-IQ3_XS.gguf) |\n| Dolphin-v2.i1-IQ3_XXS.gguf | i1-IQ3_XXS | 1.28 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-IQ3_XXS.gguf) |\n| Dolphin-v2.i1-IQ4_NL.gguf | i1-IQ4_NL | 1.83 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-IQ4_NL.gguf) |\n| Dolphin-v2.i1-IQ4_XS.gguf | i1-IQ4_XS | 1.74 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-IQ4_XS.gguf) |\n| Dolphin-v2.i1-Q2_K.gguf | i1-Q2_K | 1.27 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-Q2_K.gguf) |\n| Dolphin-v2.i1-Q2_K_S.gguf | i1-Q2_K_S | 1.2 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-Q2_K_S.gguf) |\n| Dolphin-v2.i1-Q3_K_L.gguf | i1-Q3_K_L | 1.71 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-Q3_K_L.gguf) |\n| Dolphin-v2.i1-Q3_K_M.gguf | i1-Q3_K_M | 1.59 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-Q3_K_M.gguf) |\n| Dolphin-v2.i1-Q3_K_S.gguf | i1-Q3_K_S | 1.45 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-Q3_K_S.gguf) |\n| Dolphin-v2.i1-Q4_0.gguf | i1-Q4_0 | 1.83 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-Q4_0.gguf) |\n| Dolphin-v2.i1-Q4_1.gguf | i1-Q4_1 | 2 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-Q4_1.gguf) |\n| Dolphin-v2.i1-Q4_K_M.gguf | i1-Q4_K_M | 1.93 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-Q4_K_M.gguf) |\n| Dolphin-v2.i1-Q4_K_S.gguf | i1-Q4_K_S | 1.83 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-Q4_K_S.gguf) |\n| Dolphin-v2.i1-Q5_K_M.gguf | i1-Q5_K_M | 2.22 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-Q5_K_M.gguf) |\n| Dolphin-v2.i1-Q5_K_S.gguf | i1-Q5_K_S | 2.17 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-Q5_K_S.gguf) |\n| Dolphin-v2.i1-Q6_K.gguf | i1-Q6_K | 2.54 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.i1-Q6_K.gguf) |\n| Dolphin-v2.imatrix.gguf | imatrix | 3.39 MB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.imatrix.gguf) |\n| Dolphin-v2.mmproj-Q8_0.gguf | mmproj-Q8_0 | 848 MB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.mmproj-Q8_0.gguf) |\n| Dolphin-v2.mmproj-bf16.gguf | mmproj-bf16 | 1.34 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.mmproj-bf16.gguf) |\n| Dolphin-v2.mmproj-f16.gguf | mmproj-f16 | 1.34 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.mmproj-f16.gguf) |\n| Dolphin-v2.mmproj-f32.gguf | mmproj-f32 | 2.67 GB | [Download](https://huggingface.co/prithivMLmods/Dolphin-v2-f32-GGUF/blob/main/Dolphin-v2.mmproj-f32.gguf) |\n\n## Quants Usage \n\n(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)\n\nHere is a handy graph by ikawrakow comparing some lower-quality quant\ntypes (lower is better):\n\n![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)\n",
    "related_quantizations": []
  },
  "tags": [
    "transformers",
    "gguf",
    "qwen2_5_vl",
    "text-generation-inference",
    "document-parsing",
    "document-understanding",
    "document-intelligence",
    "ocr",
    "layout-analysis",
    "table-extraction",
    "formula-recognition",
    "code-extraction",
    "vision-language-model",
    "multimodal",
    "image-text-to-text",
    "en",
    "zh",
    "base_model:ByteDance/Dolphin-v2",
    "base_model:quantized:ByteDance/Dolphin-v2",
    "license:apache-2.0",
    "endpoints_compatible",
    "region:us",
    "conversational"
  ],
  "likes": 2,
  "downloads": 339,
  "gated": false,
  "private": false,
  "last_modified": "2025-12-24T06:35:38.000Z",
  "created_at": "2025-12-24T04:58:08.000Z",
  "pipeline_tag": "image-text-to-text",
  "library_name": "transformers"
}

{
  "_id": "694b72e01c2c18ff21a17511",
  "id": "prithivMLmods/Dolphin-v2-f32-GGUF",
  "modelId": "prithivMLmods/Dolphin-v2-f32-GGUF",
  "sha": "78e5bbd5721bb730f202d645bd9a38ec8135f36a",
  "createdAt": "2025-12-24T04:58:08.000Z",
  "lastModified": "2025-12-24T06:35:38.000Z",
  "author": "prithivMLmods",
  "downloads": 339,
  "likes": 2,
  "gated": false,
  "private": false,
  "pipeline_tag": "image-text-to-text",
  "library_name": "transformers",
  "siblings_count": 46
}