# A security flaw has been discovered in vllm-project vLLM up to 0.31.0. This impacts the function get_token_bin_counts_and_mask of the file vllm/model_executor/layers/utils.py of the component Penal...

Advisory record · `PYSEC-2026-4319`

- **Identifiers:** `PYSEC-2026-4319` · `CVE-2026-105922`
- **State:** Confirmed
- **Severity:** CVSS 5.3 Medium
- **Critical:** No
- **Weakness (CWE):** [CWE-404](https://cwe.mitre.org/data/definitions/404.html)
- **Tags:** [Denial of service](https://agentprime.io/advisories/tag/dos.md)
- **Published:** 6 Oct 2026
- **Updated:** 18 Oct 2026
- **Reviewed by a person:** No

Published by OSV at 6 Oct 2026, 15:17 UTC. agentprime has not analysed this yet, so whether it affects any app is not known.

## Affected packages

- `vllm` (PyPI): affected `<0.32.0`; fixed in `0.32.0`

## Machine-written summary

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

> vLLM versions up to 0.31.0 (fixed in 0.32.0) contain a flaw in the penalty handling code, specifically the get_token_bin_counts_and_mask function. A remote attacker can manipulate input to cause a denial of service, and a public exploit exists. The source gives few details on the exact trigger. Upgrade vllm to version 0.32.0 or later.

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:L/UI:N/VC:N/VI:N/VA:L/SC:N/SI:N/SA:N/E:P`, from CNA (VulDB)

- **CISA KEV:** Not listed.
- **EPSS:** 0.0031, the 24th percentile, on 9 Nov 2026: FIRST’s estimate of the chance of exploitation in the next 30 days.

## Timeline

- 31 Oct 2026, 14:08 UTC: First seen by agentprime
- 31 Oct 2026, 14:08 UTC: [Confirmed by OSV](https://osv.dev/vulnerability/PYSEC-2026-4319)

## Sources

- [OSV: PYSEC-2026-4319](https://osv.dev/vulnerability/PYSEC-2026-4319)
- <https://github.com/vllm-project/vllm/>
- <https://gist.github.com/Yunzez/5c70a3bb328eaf6d646b9d31e879dc4a#file-repro_prompt_embeds_penalties_engine_crash-py>
- <https://vuldb.com/cve/CVE-2026-105922>
- <https://vuldb.com/submit/992707>
- <https://vuldb.com/vuln/413896>

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/PYSEC-2026-4319>, or `POST https://api.agentprime.io/v1/match` with your packages (<https://agentprime.io/api.md>)

---

The web page: <https://agentprime.io/advisories/PYSEC-2026-4319/>. Everything for agents: <https://agentprime.io/llms.txt>.