CVE-2026-85670 tokenizers BpeBuilder Buffer Overflow via merge token
- Published
- Updated
- Exploit status
- No public exploit known
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.
Affected products
| Product | Affected versions | Fixed in |
|---|---|---|
| pypi tokenizers | <= 0.23.2 | No fix listed |
References
- GitHub Issue #2094 exploit issue-tracking
- github.com product
- github.com technical-description
- VulnCheck Advisory: tokenizers BpeBuilder Buffer Overflow via merge token third-party-advisory
- Red Hat CVE page vendor-advisory
- Red Hat Bugzilla 2528822 issue-tracking
- www.cve.org
- nvd.nist.gov