CVE-2026-108955
Received Received - Intake

Out-of-Bounds Read in Tiiny-AI PowerInfer GGUF Parser

Publication date: 2026-10-13

Last updated on: 2026-10-13

Assigner: VulDB

Description
A vulnerability was determined in Tiiny-AI PowerInfer up to 8bd56d69906c9d2dba4d3bf6899763401e01a9a4. Impacted is the function gguf_init_from_file of the file ggml.c of the component GGUF Parser. This manipulation of the argument data_offset causes out-of-bounds read. The attack requires local access. The exploit has been publicly disclosed and may be utilized. This product uses a rolling release model to deliver continuous updates. As a result, specific version information for affected or updated releases is not available. The pull request to fix this issue awaits acceptance.

CVSS Scores

EPSS Scores

Probability:
Percentile:

Meta Information

Published
2026-10-13
Last Modified
2026-10-13
Generated
2026-10-13
AI Q&A
2026-10-13
EPSS Evaluated
N/A
NVD
CVE-2026-108955
EUVD
EUVD-2026-98234

Affected Vendors & Products

Showing 1 associated CPE
Vendor Product Version / Range
Tiiny-AI PowerInfer 8bd56d69906c9d2dba4d3bf6899763401e01a9a4

Helpful Resources

Exploitability

CWE
CWE Icon
KEV
KEV Icon
CWE ID Description
CWE-119 The product performs operations on a memory buffer, but it reads from or writes to a memory location outside the buffer's intended boundary. This may result in read or write operations on unexpected memory locations that could be linked to other variables, data structures, or internal program data.
CWE-125 The product reads data past the end, or before the beginning, of the intended buffer.

Attack-Flow Graph

AI Quick Actions

Instant insights powered by AI
Executive Summary

This vulnerability is an out-of-bounds read in Tiiny-AI PowerInfer's GGUF Parser. It occurs in the function gguf_init_from_file within ggml.c. The issue arises when the data_offset argument is manipulated, causing the parser to read memory outside the intended file boundaries. This happens because tensor data offsets are not validated against the actual file size when using memory-mapped files (mmap).

Detection Guidance

To detect this vulnerability, inspect GGUF model files for invalid tensor offsets. Use tools like 'file' or 'hexdump' to examine file structure. Check PowerInfer logs for crashes during model loading. Monitor for unexpected memory access patterns or crashes when loading models.

Impact Analysis

An attacker with local access could exploit this to read sensitive memory or crash the application. This may lead to information disclosure or denial-of-service. The impact is limited to local systems since the attack requires local access.

Compliance Impact

This vulnerability involves an out-of-bounds read in the GGUF model loading process, which could lead to memory disclosure or crashes. While not directly impacting GDPR or HIPAA compliance, such vulnerabilities may expose sensitive data if exploited, potentially violating confidentiality requirements under these regulations.

Mitigation Strategies

Avoid using the default mmap path for loading GGUF files. Use the non-mmap path which correctly validates tensor offsets. Disable loading of untrusted GGUF model files until a patch is applied. Monitor PowerInfer updates for an official fix.

Chat Assistant

Ask questions about this CVE
Hi! I’m here to help you understand CVE-2026-108955. Ask me anything about the vulnerability, its impact, or mitigation strategies.
0/70

EPSS Chart