CVE-2026-65315

Awaiting Analysis Awaiting Analysis - Queue

Uncontrolled Memory Allocation in Ollama GGUF Parser

Publication date: 2026-07-21

Last updated on: 2026-07-22

Assigner: VulnCheck

Description

Ollama (HEAD f0078ae) contains an uncontrolled memory allocation vulnerability in the GGUF metadata parser that allows remote attackers to crash the server by supplying a crafted GGUF file with attacker-controlled length and count fields in string lengths, tensor dimension counts, and metadata array counts that are used as allocation sizes without validation against remaining file size. Attackers can upload a sub-1KB crafted GGUF file via the blob upload and model create or pull API endpoints to trigger unrecoverable Go runtime out-of-memory fatal errors or makeslice panics that bypass recovery middleware and crash the entire server process.

CVSS Scores

EPSS Scores

Probability:
Percentile:

Meta Information

Published

2026-07-21

Last Modified

2026-07-22

Generated

2026-08-11

AI Q&A

2026-07-22

EPSS Evaluated

2026-08-09

NVD

CVE-2026-65315

EUVD

EUVD-2026-46642

Affected Vendors & Products

Vendor	Product	Version / Range
ollama	ollama	*

Helpful Resources

Exploitability

CWE

KEV

CWE ID	Description
CWE-789	The product allocates memory based on an untrusted, large size value, but it does not ensure that the size is within expected limits, allowing arbitrary amounts of memory to be allocated.

Attack-Flow Graph

Executive Summary

This vulnerability is an uncontrolled memory allocation flaw in Ollama's GGUF metadata parser. Attackers can exploit it by uploading a specially crafted GGUF file with manipulated length and count fields. These fields are used as allocation sizes without proper validation against the actual file size, causing the server to crash due to out-of-memory errors or Go runtime panics.

Detection Guidance

This vulnerability is triggered by crafted GGUF files uploaded via Ollama's API endpoints. To detect it, monitor for crashes in the Ollama server process after file uploads. Check logs for 'makeslice panic' or 'out-of-memory' errors. Inspect uploaded GGUF files for unusually large metadata or tensor dimensions.

Impact Analysis

This vulnerability allows remote attackers to crash the Ollama server by uploading a small malicious file. The crash results in a denial of service, making the server unavailable until manually restarted. It does not appear to allow data theft or code execution, but service disruption is a significant impact.

Compliance Impact

This vulnerability could indirectly impact compliance with GDPR or HIPAA by causing service disruptions or denial-of-service conditions due to server crashes. Uncontrolled memory allocation leading to crashes may result in unauthorized data access or processing interruptions, which could violate availability requirements under these regulations.

Mitigation Strategies

Update Ollama to the latest version that patches the uncontrolled memory allocation vulnerability in the GGUF metadata parser. Restrict access to the blob upload and model create or pull API endpoints to trusted users only.

Hi! I’m here to help you understand CVE-2026-65315. Ask me anything about the vulnerability, its impact, or mitigation strategies.

0/70