CVE-2026-85670: tokenizers BpeBuilder Buffer Overflow via merge token

Severity: medium (CVSS 6.5 v3.1)
Exploited in the wild: not listed in CISA KEV
EPSS: 0.5% probability of exploitation in the next 30 days
Public exploit or detection template: none known
Published: 2026-09-04 · updated 2026-09-24
Weakness: CWE-787

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

huggingface tokenizers: through 0.23.2

References

Get alerted when CVE-2026-85670 or the next vulnerability affects your stack