# CVE-2026-73558: vLLM integer overflow in kernel causing cross-user data leak

Source: https://junglewise.ai/threats/cve-2026-73558-vllm-integer-overflow-in-kernel-causing-cross-user-data-leak  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-08-13. Updated: 2026-09-25  
License: CC BY 4.0 with a link back

CVE-2026-73558 · Severity: medium · CVSS 5.3 · Published 2026-08-13

vLLM is a widely-used open-source inference engine that powers large language model deployments. An integer overflow bug in the GPU activation kernel can cause one user's AI model output to be accidentally included in another user's response when multiple requests are processed together in the same batch. An attacker can deliberately craft requests to trigger this condition and steal sensitive information from other users' AI interactions.

Technologies: [vllm (PyPI)](https://junglewise.ai/threats/technologies/pypi-vllm), [Vllm](https://junglewise.ai/threats/technologies/vllm). Vendors: [PyPI](https://junglewise.ai/threats/vendors/pypi), [Vllm](https://junglewise.ai/threats/vendors/vllm).

## Quick answers

**Is CVE-2026-73558 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-73558 being exploited in the wild.

**How severe is CVE-2026-73558?** Medium severity, with a CVSS score of 5.3, published 13 August 2026.

**Which products and versions are affected?** vllm (PyPI) ECOSYSTEM:introduced=0,fixed=0.27.0.

**Is there a fix for CVE-2026-73558?** Yes. vllm (PyPI): upgrade to 0.27.0 or later. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.

**How is CVE-2026-73558 attacked?** It is attacked over the network.

**Am I affected by CVE-2026-73558?** Check whether you run vllm (PyPI) or Vllm at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

## How to fix

- vllm (PyPI): upgrade to 0.27.0 or later

  ```sh
  pip install "vllm>=0.27.0"
  ```


## Technical details

The vulnerability is an integer overflow in the expression "blockIdx.x * 2 * d" within the act_and_mul_kernel GPU kernel (csrc/activation_kernels.cu line 82). When the intermediate computation exceeds 2^32 and wraps around to a small value (particularly when 2^32 is divisible by d, the hidden dimension size), the kernel can compute memory addresses that read from another user's input buffer instead of the current user's. This results in User A's output buffer receiving partial or complete copies of User B's inference result. The vulnerability is straightforward to trigger: using batch size 17, sequence length 16384, and d=8192 (as in Llama-3.2-1B-Instruct) causes the final batch response to be an exact copy of the first. Network-reachable vLLM services accepting multiple concurrent inference requests are vulnerable. No privilege or special user interaction is required beyond submitting crafted inference requests. The fix was released in vLLM version 0.27.0.

## Affected products

- vLLM Project vLLM < 0.27.0

## Identifiers

- CVE-2026-73558
- GHSA-7m6h-x95x-82q5
- PYSEC-2026-3935

## Timeline

- 2026-08-11: disclosed: Published to GitHub Security Advisory
- 2026-09-08: patched: Fix released in vLLM v0.27.0

## References

- [https://api.github.com/users/songlh](https://api.github.com/users/songlh)
- [https://github.com/songlh](https://github.com/songlh)
- [https://api.github.com/users/songlh/gists%7B/gist_id%7D](https://api.github.com/users/songlh/gists%7B/gist_id%7D)
- [https://api.github.com/users/songlh/repos](https://api.github.com/users/songlh/repos)
- [https://avatars.githubusercontent.com/u/1619878?v=4](https://avatars.githubusercontent.com/u/1619878?v=4)
- [https://api.github.com/users/songlh/events%7B/privacy%7D](https://api.github.com/users/songlh/events%7B/privacy%7D)

## Related threats

- [vLLM chat template resource exhaustion DoS](https://junglewise.ai/threats/cve-2026-115918-vllm-is-an-inference-and-serving-engine-for-large-language)
- [vLLM: Permissive default CORS origins allow cross-origin requests to the vLLM HTTP server](https://junglewise.ai/threats/cve-2026-115917-vllm-is-an-inference-and-serving-engine-for-large-language)
- [vLLM off-by-one error in tokenizer max_token_id validation](https://junglewise.ai/threats/cve-2026-115916-vllm-is-an-inference-and-serving-engine-for-large-language)
- [vLLM image format validation bypass in media processing](https://junglewise.ai/threats/cve-2026-115915-vllm-is-an-inference-and-serving-engine-for-large-language)
- [vLLM denial of service in DeepSeek-OCR image processing](https://junglewise.ai/threats/cve-2026-115914-vllm-is-an-inference-and-serving-engine-for-large-language)