CVE-2026-85670: Software Security Vulnerability Vulnerability

HIGH
Advisory Identifier: GHSA-jrf5-f6jp-7vxc
Share
CVSS 6.5
CWE-787
Published
Advisory Summary

CVE-2026-85670 is a HIGH (CVSS 6.5) security vulnerability vulnerability (CWE-787: Out-of-bounds Write) affecting Software libraries in the Maven ecosystem. The weakness occurs when the product writes data past the end, or before the beginning, of the intended buffer.. Remediation is available in the latest vendor maintenance releases.

CVE-2026-85670 at a Glance

CVE Identifier CVE-2026-85670
GHSA Identifier GHSA-jrf5-f6jp-7vxc
Severity HIGH
CVSS v3.1 Base Score 6.5
CWE Classification CWE-787: Out-of-bounds Write
Affected Product Software
Package Ecosystem Maven
Published Date
Last Updated

Security Impact

CVSS: 6.5

The security impact metrics below are parsed directly from the CVSS v3.1 vector string (CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H).

Attack Vector Network (AV:N)
Attack Complexity Low (AC:L)
Privileges Required None (PR:N)
User Interaction Required (UI:R)
Scope Unchanged (S:U)
Confidentiality None (C:N)
Integrity None (I:N)
Availability High (A:H)

How to Fix CVE-2026-85670

To remediate this vulnerability, upgrade all affected Software dependencies to a patched release. Both direct and transitive dependencies should be checked, as frameworks or security libraries may pull in vulnerable versions transitively.

Verify Direct and Transitive Dependencies

Run the following command in your project directory to verify that all resolved dependencies use the patched release: 

mvn dependency:tree

How to Check if You Are Affected

To determine whether your Java application is affected by CVE-2026-85670, inspect your build dependencies using the Maven Dependency Plugin:

mvn dependency:tree

Analyze the terminal output to locate any version of Software in your dependency tree. If any resolved version matches the vulnerable ranges listed in the table below (for example, versions earlier than the patched release), your application is affected.

Note: The mvn dependency:tree command inspects the resolved dependency graph declared in your project's POM files; it does not perform automated security scanning or inspect runtime dynamic classpaths. Always verify runtime container classpaths and shaded JARs.

Technical Details

Vulnerability Mechanism & Technical Analysis

This vulnerability is classified under CWE-787: Out-of-bounds Write, where the product writes data past the end, or before the beginning, of the intended buffer. In Software, software state validation or input boundaries are circumvented during processing. Review the official source advisory statement below for specific affected components, classes, and reproduction details.

Source Advisory Statement

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.

References

NVD — CVE-2026-85670 NVD
https://nvd.nist.gov/vuln/detail/CVE-2026-85670
Software Source Repository REPOSITORY
https://github.com/huggingface/tokenizers/issues/2094
Software Source Repository REPOSITORY
https://github.com/huggingface/tokenizers
Software Source Repository REPOSITORY
https://github.com/huggingface/tokenizers/blob/v0.23.2/tokenizers/src/models/bpe/model.rs
www.vulncheck.com/advisories/tokenize... WEB
https://www.vulncheck.com/advisories/tokenizers-bpebuilder-buffer-overflow-via-merge-token

Frequently Asked Questions

What is CVE-2026-85670?
CVE-2026-85670 is a security vulnerability in Software classified as a security vulnerability vulnerability under CWE-787: Out-of-bounds Write.
Which Software versions are affected?
Affected versions include releases prior to current release patches.
What version fixes CVE-2026-85670?
Upgrading to current release patches (or the latest patched release for your respective branch) resolves the vulnerability.
What is the CVSS score of CVE-2026-85670?
CVE-2026-85670 has a CVSS v3.1 base score of 6.5 (HIGH) with vector string CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H.
What CWE is associated with CVE-2026-85670?
CVE-2026-85670 is classified under CWE-787: Out-of-bounds Write (The product writes data past the end, or before the beginning, of the intended buffer.).
How can I check Maven dependencies for CVE-2026-85670?
Run the dependency tree command: 'mvn dependency:tree'. This command lists direct and transitive dependencies in your build configuration so you can verify if a vulnerable version is resolved.
Machine-readable OSV JSON OSV v1.5 Schema
Normalized schema output
View Raw JSON 
{
  "schema_version" : "1.4.0",
  "id" : "GHSA-jrf5-f6jp-7vxc",
  "modified" : "2026-09-04T15:36:15Z",
  "published" : "2026-09-04T15:36:15Z",
  "aliases" : [ "CVE-2026-85670" ],
  "details" : "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.",
  "severity" : [ {
    "type" : "CVSS_V3",
    "score" : "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H"
  }, {
    "type" : "CVSS_V4",
    "score" : "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/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X"
  } ],
  "affected" : [ ],
  "references" : [ {
    "type" : "ADVISORY",
    "url" : "https://nvd.nist.gov/vuln/detail/CVE-2026-85670"
  }, {
    "type" : "WEB",
    "url" : "https://github.com/huggingface/tokenizers/issues/2094"
  }, {
    "type" : "WEB",
    "url" : "https://github.com/huggingface/tokenizers"
  }, {
    "type" : "WEB",
    "url" : "https://github.com/huggingface/tokenizers/blob/v0.23.2/tokenizers/src/models/bpe/model.rs"
  }, {
    "type" : "WEB",
    "url" : "https://www.vulncheck.com/advisories/tokenizers-bpebuilder-buffer-overflow-via-merge-token"
  } ],
  "database_specific" : {
    "cwe_ids" : [ "CWE-787" ],
    "severity" : "HIGH",
    "github_reviewed" : false,
    "github_reviewed_at" : null,
    "nvd_published_at" : "2026-09-04T15:17:44Z"
  }
}