Skip to main content
Press slash or control plus K to focus the search. Use the arrow keys to navigate results and press enter to open a threat.
Reconnecting to live updates…

CVE-2026-75090: Out-of-Bounds Read in EricLBuehler Mistral.rs

0
Medium
VulnerabilityCVE-2026-75090cvecve-2026-75090
Published: 08/18/2026 (08/18/2026, 01:15:14 UTC)
Source: CVE Database V5
Vendor/Project: EricLBuehler
Product: Mistral.rs

Description

CVE-2026-75090 is an out-of-bounds read vulnerability in the EricLBuehler Mistral.rs library up to version 0.8.22. The flaw exists in the convert_gguf_to_hf_tokenizer function of the GGUF Tokenizer component and involves manipulation of eos_token_id, bos_token_id, or unknown_token_id arguments. This vulnerability can be exploited remotely and public exploit code is available. Upgrading to version 0.8.23 resolves the issue.

CVSS v4.0

Score 5.3medium

Attack Vector
Network
Attack Complexity
Low
Attack Requirements
None
Privileges Required
None
User Interaction
Passive
Vuln. Confidentiality
None
Vuln. Integrity
None
Vuln. Availability
Low
Subsq. Confidentiality
None
Subsq. Integrity
None
Subsq. Availability
None
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:P/VC:N/VI:N/VA:L/SC:N/SI:N/SA:N/E:P

Affected software

crates.iomore threats →ai
github/EricLBuehler/mistral.rs
pkg:cargo/github/EricLBuehler/mistral.rs
Affected versions
=0.8.0=0.8.1=0.8.2=0.8.3=0.8.4=0.8.5=0.8.6=0.8.7=0.8.8=0.8.9=0.8.10=0.8.11=0.8.12=0.8.13=0.8.14=0.8.15=0.8.16=0.8.17=0.8.18=0.8.19=0.8.20=0.8.21=0.8.22
CPE configurations
cpe:2.3:a:ericlbuehler:mistral.rs:*:*:*:*:*:*:*:*

Run on your own infrastructure? Check whether these packages are installed with threat-finder — our free open-source scanner.

AI-Powered Analysis

Machine-generated threat intelligence

AILast updated: 08/18/2026, 02:16:12 UTC

Technical Analysis

The vulnerability CVE-2026-75090 affects the EricLBuehler Mistral.rs project, specifically versions up to and including 0.8.22. It is located in the convert_gguf_to_hf_tokenizer function within the mistralrs-core/src/gguf/gguf_tokenizer.rs file. The issue arises from improper handling of the eos_token_id, bos_token_id, and unknown_token_id arguments, leading to an out-of-bounds read condition. This flaw can be triggered remotely without privileges or user interaction, and public exploit code exists. The vulnerability is fixed in version 0.8.23, with the patch identified by commit cd5297e2ea5cb27c790bdcf2f3c2f1064a81d55e.

Potential Impact

An attacker can remotely trigger an out-of-bounds read in the GGUF Tokenizer component by manipulating specific token ID arguments. This may lead to information disclosure or application instability. The CVSS 4.0 base score is 5.3 (medium severity), reflecting network attack vector, low complexity, no privileges required, and limited impact on confidentiality.

Mitigation Recommendations

Upgrade EricLBuehler Mistral.rs to version 0.8.23 or later, which contains the official fix for this vulnerability. No other mitigations are indicated by the vendor advisory.

Pro Console: star threats, build custom feeds, automate alerts via Slack, email & webhooks.Upgrade to Pro

Technical Details

Data Version
5.2
Assigner Short Name
VulDB
Date Reserved
2026-08-17T16:35:41.582Z
Cvss Version
4.0
State
PUBLISHED
Remediation Level
null

Threat ID: 6a83bbd9bf8831d539fd23f5

Added to database: 08/18/2026, 01:56:41 UTC

Last enriched: 08/18/2026, 02:16:12 UTC

Last updated: 08/18/2026, 02:16:12 UTC

Views: 5

Community Reviews

0 reviews

Crowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.

Sort by
Loading community insights…

Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.

Actions

PRO

Updates to AI analysis require Pro Console access. Upgrade inside Console → Billing.

Please log in to the Console to use AI analysis features.

Need more coverage?

Upgrade to Pro Console for AI refresh and higher limits.

For incident response and remediation, OffSeq services can help resolve threats faster.

Latest Threats

Breach by OffSeqOFFSEQFRIENDS — 25% OFF

Check if your credentials are on the dark web

Instant breach scanning across billions of leaked records. Free tier available.

Scan now
OffSeq TrainingCredly Certified

Lead Pen Test Professional

Technical5-day eLearningPECB Accredited
View courses