INCIDENT MAP

Browse / Model Evasion / CVE-2026-12491

vLLM: image EXIF Rotation & PNG tRNS Transparency Not Normalized, Causing Mismatch Between Model Input and Expectations

CVE-2026-12491 · 2026-06-17 moderate

Class
Model Evasion (filed by a model that read the advisory)
Technique
Other weaknesses
Affected
vllm (PyPI)
CVSS
CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:L/A:L
EPSS
0.2% chance of exploitation in 30 days

How it works, in short

Images with EXIF rotation or PNG transparency tricks are preprocessed differently than humans see them, so the model receives hidden or altered content. (Our summary, written by a model. The source text is below.)

Details

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

---

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

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

References

Source: GitHub Advisory Database, used under CC BY 4.0.