CVE-2026-56340: Vllm Improper Input Validation (CVSS 8.7)

vLLM versions 0.22.0 through 0.23.0 fail to validate stop_token_ids against vocabulary bounds in Rust HTTP and gRPC frontends, allowing out-of-vocabulary token IDs to reach MinTokensLogitsProcessor. Attackers can submit requests with min_tokens greater than zero and out-of-vocabulary stop_token_ids to trigger CUDA tensor indexing failures that leave EngineCore in a fatal state requiring service restart.

vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Prior to 0.24.0, the structured_outputs.regex API parameter passes a user-supplied regular expression string directly to the grammar compiler backends with no compilation timeout; in the xgrammar backend the string reaches the regex compiler with no guard, and in the outlines backend the validation step blocks structural issues such as lookarounds and backreferences but performs no complexity analysis, so a pattern with nested quantifiers passes all checks and causes exponential state-space expansion, allowing a single request containing an adversarial regex to hang an inference worker indefinitely and deny service. This issue is fixed in version 0.24.0.

vLLM is an inference and serving engine for large language models (LLMs). The SSRF protection fix for CVE-2026-24779 add in 0.15.1 can be bypassed in the load_from_url_async method due to inconsistent URL parsing behavior between the validation layer and the actual HTTP client. The SSRF fix uses urllib3.util.parse_url() to validate and extract the hostname from user-provided URLs. However, load_from_url_async uses aiohttp for making the actual HTTP requests, and aiohttp internally uses the yarl library for URL parsing. This vulnerability in 0.17.0.

vLLM is an inference and serving engine for large language models (LLMs). Starting in version 0.10.1 and prior to version 0.14.0, vLLM loads Hugging Face `auto_map` dynamic modules during model resolution without gating on `trust_remote_code`, allowing attacker-controlled Python code in a model repo/path to execute at server startup. An attacker who can influence the model repo/path (local directory or remote Hugging Face repo) can achieve arbitrary code execution on the vLLM host during model load. This happens before any request handling and does not require API access. Version 0.14.0 fixes the issue.

Improper Neutralization used in an OS Command in the container launcher in Google Gemini CLI (versions prior to 0.39.1) and run-gemini-cli GitHub Action (versions prior to 0.1.22) on headless CI platforms allows an unprivileged attacker to achieve pre-sandbox host-level code execution a maliciously crafted .gemini/.env file.

Budibase is a low code platform for creating internal tools, workflows, and admin panels. Prior to version 3.30.4, an unsafe `eval()` vulnerability in Budibase's view filtering implementation allows any authenticated user (including free tier accounts) to execute arbitrary JavaScript code on the server. This vulnerability ONLY affects Budibase Cloud (SaaS) - self-hosted deployments use native CouchDB views and are not vulnerable. The vulnerability exists in `packages/server/src/db/inMemoryView.ts` where user-controlled view map functions are directly evaluated without sanitization. The primary impact comes from what lives inside the pod's environment: the `app-service` pod runs with secrets baked into its environment variables, including `INTERNAL_API_KEY`, `JWT_SECRET`, CouchDB admin credentials, AWS keys, and more. Using the extracted CouchDB credentials, we verified direct database access, enumerated all tenant databases, and confirmed that user records (email addresses) are readable. Version 3.30.4 contains a patch.