Your RSA-2048 keys break in 2030. Find every one of them before attackers do.
🐍
🐍 PyPI
Not in CISA KEV
MEDIUM severity

CVE-2026-12491 — vllm

MEDIUMFix: vllm-project/vllm#44974

CVE-2026-12491 is a medium-severity (CVSS 4.8) CWE-115 vulnerability in vllm. A fix is available for vllm — see the affected versions and patch details below.

Vllm: vllm: image exif rotation & png trns transparency not normalized, causing mismatch between model input and expectations

Also known asGHSA-8jr5-v98p-w75mPYSEC-2026-3406
Published
Updated
Affected
1 pkg
Patched
1 / 1
Exploits
None indexed
Exploitation data as of Oct 23, 2026 · OSV.dev, NVD, FIRST.org (EPSS)

Exploitation Status

No confirmed exploitation observed yet

  • CISA’s own triage has not observed active exploitation or public proof-of-concept code for this CVE as of its last assessment.

Exploitation and automatability from CISA’s SSVC triage for CVE-2026-12491.

EPSS Exploitation Probability

via FIRST.org ↗
0.2%probability of exploitation in next 30 days
Lower Risk0.00%
Lower risk than most CVEs14th percentile — riskier than 14% of all scored CVEsHighest risk
0.00%0.25%0.49%0.74%0.2%0.2%0.2%0.2%0.2%Jul 26Sep 26Oct 26

Probability of exploitation in the next 30 days, from FIRST.org EPSS.

How urgent is this, really

CVE-2026-12491 by exploitation likelihood (EPSS) against impact (CVSS). Outside the shaded patch-first corner.

Where this sits among everything scored

Of 394,803 CVEs with a current EPSS score, this one falls in the < 10% band (highlighted). Counts from FIRST.org, log-scaled.

Real-World Exposure

1 pkg affected
🐍vllm

Real-time download stats are indexed for npm and PyPI packages. This vulnerability affects PyPI packages — download data is not available via public APIs for these ecosystems.

Description

Summary

Issue 1: EXIF orientation not normalized → The image orientation processed by the model differs from how humans view it, introducing interpretation bias.

Issue 2: PNG tRNS not explicitly flattened before converting to RGB → After conversion, transparent/semi-transparent pixels are rendered unexpectedly, making otherwise subtle overlay elements visible and distorting the input content. (This attack is similar to AlphaDog: RGBA handling is already correct in vLLM, but since tRNS permits RGB images, the correct processing path isn’t taken.)

Issue 3 : Pillow only loads the first frame when loading APNG or GIF files.

Root Cause

  • Rotation: After opening an image, ImageOps.exif_transpose is not called to normalize EXIF orientation.
  • Transparency: Only RGBA→RGB is flattened with a background; PNGs carrying tRNS in P/L/RGB + tRNS and other non-RGBA modes take the image.convert("RGB") path, which implicitly discards/remaps transparency semantics.

Affected Code

https://github.com/vllm-project/vllm/blob/16b37f3119918c1e5a39f303e0d0892c65c07a90/vllm/multimodal/image.py#L77-L84

https://github.com/vllm-project/vllm/blob/16b37f3119918c1e5a39f303e0d0892c65c07a90/vllm/multimodal/image.py#L37-L43

https://github.com/vllm-project/vllm/blob/16b37f3119918c1e5a39f303e0d0892c65c07a90/vllm/multimodal/image.py#L26-L34

Current state: ImageOps.exif_transpose is not used. (Although the rescale_image_size function (https://github.com/vllm-project/vllm/blob/main/vllm/multimodal/image.py#L14) exists and includes a transpose parameter, I’ve found that it doesn’t seem to be called anywhere outside the test directory.）

Call order: _convert_image_mode runs first; if the conditions are met, convert_image_mode is called.

Issue: Only the “RGBA → RGB” path is explicitly flattened. P, L, or RGB with tRNS all fall back to image.convert("RGB"). For PNGs that include tRNS, convert("RGB") directly produces 24-bit RGB, leading to:

  • P mode: The transparent index becomes an actual RGB color (often black, white, or an undefined background), so transparency is lost.
  • L/LA and RGB + tRNS: convert("RGB") doesn’t composite against a chosen background first, so elements that relied on transparency to be hidden or softened become solid.

Impact & Scope

  • Impact: Pixels the model sees can diverge from operator expectations (due to orientation or transparency handling), potentially altering downstream reasoning.
  • Scope: The image I/O and mode-conversion paths in vllm/multimodal/image.py. The existing RGBA→RGB flattening is correct; the issues center on missing EXIF normalization and non-RGBA tRNS not being explicitly composited.

Case

EXIF： http://qiniu.funxingzuo.top/exif_orient_180.jpg tRNS: http://qiniu.funxingzuo.top/hello.png

Fix

A fix for this vulnerability was merged here: https://github.com/vllm-project/vllm/pull/44974

Affected Packages

1 total 1 fixed
EcosystemPackageVulnerable rangeFix
🐍PyPIvllm≥ 0.11.0&&< 0.24.00.24.0pip install --upgrade 'vllm==0.24.0'

Detection & mitigation playbook

Open-source dependency

  1. Detect

    Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for vllm, including transitive dependencies — a direct dependency you never call can still pull in a vulnerable version.

  2. Fix

    Update vllm to 0.24.0 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms CVE-2026-12491 is resolved across your whole dependency graph.

  3. Workarounds

    If you can't upgrade right away: gate or disable the affected feature, validate untrusted input at the boundary, and avoid passing attacker-controlled data into the vulnerable path. O3's runtime protection blocks exploitation in production as an interim safeguard until the upgrade lands.

Frequently Asked Questions

## Summary Issue 1: EXIF orientation not normalized → The image orientation processed by the model differs from how humans view it, introducing interpretation bias. Issue 2: PNG tRNS not explicitly flattened before converting to RGB → After conversion, transparent/semi-transparent pixels are rendered unexpectedly, making otherwise subtle overlay elements visible and distorting the input content. (This attack is similar to AlphaDog: RGBA handling is already correct in vLLM, but since tRNS permits RGB images, the correct processing path isn’t taken.) Issue 3 : Pillow only loads the first fram
O3 Security · Impact-Aware SCA

Is CVE-2026-12491 in your dependencies?

Find it across PyPI, including transitive dependencies.

Scan my dependencies How O3 SCA works
CVE-2026-12491: vllm — Fixed in 0.24.0