CVE-2026-22778
Official description Straight from the sourceThe vendor's or NVD's own wording, published unedited. Authoritative, but often terse — it says what broke, rarely what to do.
NVD · uneditedvLLM is an inference and serving engine for large language models (LLMs). From 0.8.3 to before 0.14.1, when an invalid image is sent to vLLM's multimodal endpoint, PIL throws an error. vLLM returns this error to the client, leaking a heap address. With this leak, we reduce ASLR from 4 billion guesses to ~8 guesses. This vulnerability can be chained a heap overflow with JPEG2000 decoder in OpenCV/FFmpeg to achieve remote code execution. This vulnerability is fixed in 0.14.1.
Technical summary Written by usOur analysis, written from the advisory, the CVSS vector and the affected-version data. It adds context the advisory leaves out, and never invents facts that are not in the source.
dbcve analysis · high confidencevLLM versions 0.8.3 through before 0.14.1 leak heap addresses through PIL error messages when processing invalid images at the multimodal endpoint. This information disclosure breaks ASLR by reducing address entropy from 4 billion to approximately 8 possible guesses, enabling reliable exploitation of a separate heap overflow vulnerability in the JPEG2000 decoder (OpenCV/FFmpeg) for remote code execution.
Verify against the referenced sources before acting — the references below are authoritative for this CVE, this summary is not.
Affected products & versions What the vendor confirmedThe version ranges the vendor confirmed as vulnerable. If your version sits inside a range here, treat yourself as exposed until you have upgraded.
NVD · CPE data>= 0.8.3, < 0.14.1CVSS breakdown How the score is builtThe industry scoring standard. It rates how the flaw is reached, what it takes to exploit, and what an attacker gains — the score is derived from those, not the other way round.
From the vector- Attack vector
- Network
- Complexity
- Low
- Privileges
- None
- User interaction
- None
- Scope
- Unchanged
- Confidentiality
- High
- Integrity
- High
- Availability
- High
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
Am I affected? How to checkSteps we derive from the advisory and the affected-version data, so you can decide whether this CVE reaches your setup. They are a guide, not a scan — your own configuration is the authority.
dbcve checksWork through these to decide whether this CVE applies to you.
-
Identify installed vLLM versionRun `pip show vllm` or check the version from your deployment environmentAffected if Version is >= 0.8.3 and < 0.14.1
-
Confirm multimodal endpoint is activeCheck if the vLLM server was started with multimodal model support (e.g., using --multimodal-mounts or loading a vision-capable model like LLaVA, Qwen2-VL, or similar)Affected if Multimodal inference endpoint is enabled and accessible
-
Verify network exposure of inference APIInspect the server launch configuration or network binding (check if --host is set to a public IP or if the service is behind an unrestricted network)Affected if The multimodal API is exposed to untrusted network segments (not localhost-only or behind authentication)
You are affected if vLLM version is between 0.8.3 and 0.14.1 (exclusive), multimodal endpoint is enabled, and the API is network-accessible to send malformed images that trigger the PIL error messages.
Generated from the published advisory. Verify against your own configuration.
Remediation Closing itWhat it takes to close this. Where a vendor fix exists we point at it; where none exists we say so plainly, and can build one. Effort estimates are scoped from the advisory, not from your codebase.
dbcve · scoped0.14.1
Upgrade vLLM to version 0.14.1 or later. Until upgrade, restrict access to the multimodal API endpoint to trusted clients only.
0.14.1
- 1. Identify the current vLLM version in use by checking the installed package (e.g., `pip show vllm` or `python -c "import vllm; print(vllm.__version__)"`).
- 2. Ensure any running vLLM server instances are stopped to prevent downtime during the upgrade.
- 3. Upgrade vLLM to version 0.14.1 or later using pip: `pip install --upgrade vllm==0.14.1` or `pip install --upgrade 'vllm>=0.14.1'` to get the latest fixed version.
- 4. Verify the upgrade was successful by checking the installed version again.
- 5. Restart the vLLM service and test that multimodal image processing still functions correctly with the fix applied.
Generated from the published advisory — verify against the referenced sources before acting.
- Consultation4.0 h
- Implementation4.0 h
- Testing6.0 h
- Review / QA3.0 h
An estimate, not a bill — we confirm scope with you before any work starts. Need it this week? Rush from $4,736.
Scan for this in your stack
Free · runs locallyCheck whether your project pulls in CVE-2026-22778 — or any other known-vulnerable package — straight from your lock files. Free and open source; it runs locally and uploads nothing.
References Go to the primary sourcePrimary sources — vendor advisories, patches and trackers. Where our summary and a reference disagree, the reference wins.
Primary sourcesPractitioner notes
ContributedPeer-ranked notes from engineers who’ve handled CVE-2026-22778 in production — separate from our analysis above.
The advisory tells you what broke. It rarely tells you what actually worked. If you’ve dealt with this one, that detail is what the next engineer is searching for.
- The version that genuinely resolved it — not the one the vendor claimed
- A config change or rule that shut the vector down
- A gotcha in the upgrade path that cost you an afternoon
No notes yet
Be the first to add a field note for this CVE — a mitigation you’ve verified, a version caveat, or a link to a working fix. Sign in above to contribute.
A place for practitioners to share what actually worked: a mitigation you’ve tested, a configuration change, a version- or environment-specific caveat, or a link to a verified patch. The most useful notes rise to the top as peers upvote them, so the signal stays high.
- Verified mitigations, workarounds, and config changes
- Version or environment caveats, and links to real fixes
- No weaponised exploit code, or anything meant to cause harm
- No spam, self-promotion, credentials, or personal data