Ollama: Unauthenticated remote information disclosure vulnerability in Ollama's model quantization…
Ollama
More for Ollama RSS feed for Ollama
- Severity
- High, CVSS 7.5
- CVSS vector
- CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N
- Record
- CVE-2026-5757
- Affects
- Ollama
- Known exploited
- Not listed in the CISA Known Exploited Vulnerabilities catalog
- Published
- Record first published by the source
- June 26, 2026
- Source last modified
- October 25, 2026
Description
Unauthenticated remote information disclosure vulnerability in Ollama's model quantization engine allows an attacker to read and exfiltrate the server's heap memory, potentially leading to sensitive data exposure, further compromise, and stealthy persistence.
Description quoted from the NVD record.
Who is affected
Ollama, a program that runs AI language models on your own computer or server, has a flaw in the part that makes AI models smaller so they use less space and memory. No specific affected versions are named for this flaw. An attacker who can reach the Ollama server over a network, without logging in, could read and copy information held in the server's working memory. This could expose sensitive data, help the attacker break in further, and let them keep hidden access over time. This flaw is rated High.
What to do
- Whoever installed or runs Ollama in your organization (a staff member, volunteer, or IT provider) should handle this flaw.
- Find out which computers or servers in your organization run Ollama, so none are missed when the fix is installed.
- No fixed version is named for this flaw yet, so check the vendor advisory linked on this page for a fixed release and install it as soon as it is available.
- If an IT provider or vendor runs Ollama for your organization, send them this advisory and ask when the update will be done.
Analysis written from the source record as of October 25, 2026 and reviewed before publishing.