CVE-2026-114053High· 8.1▾ TwilightvLLM from 0.7.3 before 0.31.0 forwards the request-supplied mm_processor_kwargs.code_revision to Transformers AutoProcessor.from_pretrained, letting clients select which revision of a model's dynamic processor code is imported when trust…
▾ Twilight zone — High severity, or a signal on a lesser flaw
impact 44.6 · likelihood 0 · exploitation 0
Need a working PoC? Pro members can cast a request and our team develops one — it lands right here.
vLLM from 0.7.3 before 0.31.0 forwards the request-supplied mm_processor_kwargs.code_revision to Transformers AutoProcessor.from_pretrained, letting clients select which revision of a model's dynamic processor code is imported when trust_remote_code is enabled. An attacker controlling a revision in the served model repository can send a multimodal request to the unauthenticated /tokenize endpoint to execute arbitrary Python in the vLLM API process.
vllm >= 0.7.3 < 0.31.0Refer to the linked advisories for vendor-supplied fixes and affected version ranges.
Connected by shared product, vendor, weakness, or advisory.
CVE-2026-90553High· 7.8vLLM before 0.28.0 contains a remote code execution vulnerability in the LlavaOnevision2 processor loader that ignores the trust_remote_code parameter when loading remote processor classes
CVE-2026-22807High· 8.8vLLM is an inference and serving engine for large language models (LLMs)
CVE-2026-114049Low· 3.7vLLM before 0.31.0 Stale allowed_token_ids Mask via InputBatch.condense
CVE-2026-114052High· 7.5vLLM 0.25.0 before 0.31.0 MOSS-Audio Memory Exhaustion via /tokenize mm_processor_kwargs
CVE-2026-114051High· 7.5vLLM before 0.31.0 Denial of Service via min_tokens in /inference/v1/generate
CVE-2026-114050High· 7.5vLLM before 0.31.0 Remote Engine Crash via Forged Multimodal UUID