{
  "id": "CVE-2026-85670",
  "url": "https://poc.griffinguard.io/cve/CVE-2026-85670",
  "published_at": "2026-09-04T14:32:25.391Z",
  "title": "tokenizers BpeBuilder Buffer Overflow via merge token",
  "vendor": "huggingface",
  "product": "tokenizers",
  "cvss": {
    "score": 7.1,
    "source": "cna",
    "vector": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:P/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N"
  },
  "kev": {
    "in_kev": false,
    "added_at": null,
    "days_after_publication": null
  },
  "first_poc": null,
  "poc_repo_count": 0,
  "exploit_count": 0,
  "suspicious_repo_count": 0,
  "edge_device": null,
  "description": "tokenizers (Hugging Face) is affected by an out-of-bounds buffer access in BpeBuilder::build (tokenizers/src/models/bpe/model.rs). When loading a tokenizer.json via Tokenizer::from_file/from_str, the builder sizes a scratch buffer to the longest vocabulary key, then writes each concatenated merge rule into it. A merge whose concatenated token exceeds the longest vocabulary key overruns the buffer, which Rust turns into a panic that aborts the process in Rust and FFI embeddings. This occurs at load time with no encoding required, so an attacker who supplies a crafted tokenizer.json can cause a denial of service. A secondary defect at the same location can cause a usize underflow (panic in debug, potential memory corruption in release) when continuing_subword_prefix is set and a merge token is shorter than the prefix. Observed in version 0.23.1.",
  "cwe": "CWE-787",
  "cvss_by_source": {
    "cna": 7.1
  },
  "repos": [],
  "remediation": {
    "affected": [
      {
        "vendor": "huggingface",
        "product": "tokenizers",
        "ranges": [
          "\u003c= 0.23.2"
        ],
        "fixed_in": null
      }
    ]
  },
  "exploit_refs": [],
  "hunt_profile": [],
  "rules": []
}
