CVE-2026-17501
Received Received - Intake

Remote Resource Allocation in llama.cpp JSON-Schema Conversion

Publication date: 2026-07-27

Last updated on: 2026-07-27

Assigner: VulDB

Description
A flaw has been found in ggml-org llama.cpp e15efe0. This vulnerability affects the function transform of the file common/json-schema-to-grammar.cpp of the component JSON-Schema-to-GBNF Conversion. This manipulation causes allocation of resources. The attack may be initiated remotely. The pull request to fix this issue awaits acceptance.

CVSS Scores

EPSS Scores

Probability:
Percentile:

Meta Information

Published
2026-07-27
Last Modified
2026-07-27
Generated
2026-07-27
AI Q&A
2026-07-27
EPSS Evaluated
N/A
NVD
CVE-2026-17501
EUVD
EUVD-2026-49061

Affected Vendors & Products

Showing 1 associated CPE
Vendor Product Version / Range
ggml-org llama.cpp From e15efe0 (inc)

Helpful Resources

Exploitability

CWE
CWE Icon
KEV
KEV Icon
CWE ID Description
CWE-400 The product does not properly control the allocation and maintenance of a limited resource.
CWE-770 The product allocates a reusable resource or group of resources on behalf of an actor without imposing any intended restrictions on the size or number of resources that can be allocated.

Attack-Flow Graph

AI Quick Actions

Instant insights powered by AI
Executive Summary

This vulnerability is a stack overflow flaw in the JSON-Schema-to-GBNF conversion process within the llama.cpp project. It occurs due to uncontrolled recursion when processing deeply nested schemas, where the converter recursively handles nested constructs without enforcing a depth limit. This can lead to stack exhaustion and crash the process.

Detection Guidance

To detect this vulnerability, monitor for crashes or stack overflow errors in applications using llama.cpp's json-schema-to-grammar component. Check for SIGSEGV errors or process hangs when processing JSON schemas. Use tools like gdb to trace stack usage during schema processing. Validate if your llama.cpp version is affected by testing with deeply nested schemas.

Impact Analysis

The vulnerability can cause crashes by exhausting the stack when processing malformed or deeply nested JSON schemas. It may allow remote attackers to disrupt services by sending specially crafted inputs that trigger the stack overflow or null dereference errors.

Compliance Impact

This vulnerability does not directly affect compliance with GDPR or HIPAA as it involves a stack overflow in JSON-Schema-to-GBNF conversion due to uncontrolled recursion. However, if exploited, it could lead to service disruption or crashes in systems processing user-supplied schemas, potentially impacting availability of services handling sensitive data.

Mitigation Strategies

Immediately update to the latest version of llama.cpp that includes the fix for CVE-2026-17501. If updating is not possible, implement a maximum recursion depth limit in the json-schema-to-grammar component. Reject any JSON schemas that exceed this depth threshold. Add input validation to check for malformed regex patterns before processing.

Chat Assistant

Ask questions about this CVE
Hi! I’m here to help you understand CVE-2026-17501. Ask me anything about the vulnerability, its impact, or mitigation strategies.
0/70

EPSS Chart