# vLLM: GGUF dequantize kernel int truncation exposes uninitialized GPU memory in multi-tenant serving

Advisory record · `GHSA-5jv2-g5wq-cmr4`

- **Identifiers:** `GHSA-5jv2-g5wq-cmr4` · `CVE-2026-53923` · `PYSEC-2026-3403`
- **State:** Confirmed
- **Severity:** CVSS 5.3 Medium
- **Critical:** No
- **Weakness (CWE):** [CWE-200](https://cwe.mitre.org/data/definitions/200.html) · [CWE-681](https://cwe.mitre.org/data/definitions/681.html)
- **Tags:** [Information disclosure](https://agentprime.io/advisories/tag/info-disclosure.md) · [CWE-681](https://agentprime.io/advisories/tag/cwe-681.md)
- **Published:** 17 Jun 2026
- **Updated:** 10 Sept 2026
- **Reviewed by a person:** No

Published by the GitHub Advisory Database at 17 Jun 2026, 14:03 UTC. agentprime has not analysed this yet, so whether it affects any app is not known.

## Affected packages

- `vllm` (PyPI): affected `>=0.5.5, <0.24.0`; fixed in `0.24.0`

## Machine-written summary

Summary written by Claude Haiku from the source advisory. Not reviewed by a person.

> Affected: vLLM versions from 0.5.5 before 0.24.0 (PyPI vllm). Integer truncation in the GGUF dequantize CUDA kernels can leave part of a tensor's output buffer unwritten, so it retains stale GPU memory that may contain data from other users' requests in multi-tenant serving. An attacker who supplies a crafted GGUF model file with very large tensor dimensions could cause this when a victim loads the model, leading to information disclosure. Upgrade to vLLM 0.24.0 or later. The source gives no workaround.

Written 1 Nov 2026.

## Not analysed yet

We haven’t published our analysis of this Advisory yet. Once a person has reviewed it, this page will say what’s vulnerable, what has to be true for your app to be affected, and how to check.

## Severity and exploitation

- 5.3 Medium, CVSS 4.0: `CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:P/VC:L/VI:L/VA:N/SC:N/SI:N/SA:N`, from CNA (GitHub_M)

- **CISA KEV:** Not listed.
- **EPSS:** 0.0048, the 41st percentile, on 5 Nov 2026: FIRST’s estimate of the chance of exploitation in the next 30 days.

## Timeline

- 31 Oct 2026, 14:11 UTC: First seen by agentprime
- 31 Oct 2026, 14:11 UTC: [Confirmed by OSV](https://osv.dev/vulnerability/GHSA-5jv2-g5wq-cmr4)

## Sources

- [GitHub Advisory Database: GHSA-5jv2-g5wq-cmr4](https://github.com/advisories/GHSA-5jv2-g5wq-cmr4)
- [OSV: GHSA-5jv2-g5wq-cmr4](https://osv.dev/vulnerability/GHSA-5jv2-g5wq-cmr4)
- <https://github.com/vllm-project/vllm/security/advisories/GHSA-5jv2-g5wq-cmr4>
- <https://nvd.nist.gov/vuln/detail/CVE-2026-53923>
- <https://github.com/vllm-project/vllm/pull/44971>
- <https://github.com/vllm-project/vllm/commit/f219788f91952827132fa4fdf916427cd20d225e>
- <https://github.com/advisories/GHSA-5jv2-g5wq-cmr4>
- <https://github.com/pypa/advisory-database/tree/main/vulns/vllm/PYSEC-2026-3403.yaml>
- <https://github.com/vllm-project/vllm>
- <https://pypi.org/project/vllm>

Facts on this page come from the GitHub Advisory Database (CC-BY 4.0), OSV and the other sources listed, as published. We don’t reproduce full advisory descriptions, which can contain exploit code.

## Check whether this affects you

A Match means an installed version is in the affected range above. Most don’t affect the app; agentprime hasn’t analysed this one yet.

- With the agentprime MCP server (<https://api.agentprime.io/mcp>), ask your agent to check the project: `check_lockfile` sends a lockfile, `check_packages` a list of installed packages, and both return the Matches and the upgrades that clear them. `get_advisory` returns this record. Setup: <https://agentprime.io/mcp.md>
- In a browser: <https://agentprime.io/check/> reads your lockfile in the browser, never uploading it.
- Whether it touches your code, with file and line: the Claude Code plugin, <https://agentprime.io/plugin.md>
- As JSON: <https://api.agentprime.io/v1/advisories/GHSA-5jv2-g5wq-cmr4>, or `POST https://api.agentprime.io/v1/match` with your packages (<https://agentprime.io/api.md>)

---

The web page: <https://agentprime.io/advisories/GHSA-5jv2-g5wq-cmr4/>. Everything for agents: <https://agentprime.io/llms.txt>.