{
  "added_at": "2026-11-06",
  "additional_evidence_urls": [
    "https://github.com/mjbommar/glaurung-models/blob/1819020f447d55f55b629b78ba96502edae71715/SSE_AVX_ANALYSIS.md"
  ],
  "category": "native-analysis",
  "description_en": "Glaurung Models contains tokenizer and model experiments for executable analysis. Its documented 64K BPE tokenizer encodes compiled PE and ELF bytes using Latin-1, with x86 and ARM64 training coverage; an accompanying analysis compares learned SSE and AVX instruction patterns across two tokenizers.",
  "description_zh": "Glaurung Models 收录面向可执行文件分析的分词器与模型实验。其 64K BPE 分词器以 Latin-1 编码 PE 和 ELF 字节，训练范围涵盖 x86 与 ARM64；附带分析比较了两个分词器学到的 SSE 和 AVX 指令模式。",
  "entry_url": "https://awesomerea.org/resources/glaurung-models/",
  "entry_url_zh": "https://awesomerea.org/zh/resources/glaurung-models/",
  "evidence_url": "https://github.com/mjbommar/glaurung-models/blob/1819020f447d55f55b629b78ba96502edae71715/glaurung_binary_tokenizer_001_README.md",
  "group": "learn",
  "id": "glaurung-models",
  "json_url": "https://awesomerea.org/resources/glaurung-models/record.json",
  "kind": "research",
  "markdown_url": "https://awesomerea.org/resources/glaurung-models/index.md",
  "markdown_url_zh": "https://awesomerea.org/zh/resources/glaurung-models/index.md",
  "name": "Glaurung Models",
  "name_zh": "Glaurung Models 可执行文件分词模型研究",
  "record_type": "resource",
  "relationship": "learning",
  "requirement_facets": [
    "prerequisites",
    "testing",
    "license"
  ],
  "requirements": [
    "The documented tokenizer workflow uses the tokenizers and transformers Python packages and Latin-1 conversion of binary bytes.",
    "The document reports an average of 2.849 bytes per token for the 64K tokenizer.",
    "No licence file in the repository."
  ],
  "requirements_zh": [
    "文档中的分词流程使用 Python 的 tokenizers 和 transformers 包，并将二进制字节转换为 Latin-1。",
    "文档报告该 64K 分词器平均每个 token 对应 2.849 字节。",
    "仓库中没有许可证文件。"
  ],
  "targets": [
    "binary"
  ],
  "url": "https://github.com/mjbommar/glaurung-models",
  "verification": "source-reviewed",
  "verified_at": "2026-11-06"
}