CVE-2026-115916Medium· 6.5▾ SunlitvLLM is an inference and serving engine for large language models. Prior to 0.32.0, vllm/tokenizers/hf.py computes max_token_id with an off-by-one upper bound, allowing a pre-tokenized prompt ID equal to the vocabulary size to pass valid…
▾ Sunlit zone — Low / medium · no exploitation signal
impact 35.8 · likelihood 0 · exploitation 0
Need a working PoC? Pro members can cast a request and our team develops one — it lands right here.
vLLM is an inference and serving engine for large language models. Prior to 0.32.0, vllm/tokenizers/hf.py computes max_token_id with an off-by-one upper bound, allowing a pre-tokenized prompt ID equal to the vocabulary size to pass validation. For a tensor-parallel size of one and an exact multiple-of-64 vocabulary, VocabParallelEmbedding.forward does not mask that value before F.embedding uses it, causing a CUDA device-side assertion and terminating the engine. This issue is fixed in version 0.32.0.
vllm < 0.32.0Refer to the linked advisories for vendor-supplied fixes and affected version ranges.
Connected by shared product, vendor, weakness, or advisory.
CVE-2026-115911Medium· 6.5vLLM: DoS via unbounded chat_template Jinja2 rendering exhausts the renderer thread pool
CVE-2026-115912High· 7.5vLLM: Public multimodal SHM handle forgery reaches worker-side `pickle.loads()` with insecure serialization disabled
CVE-2026-115913Medium· 6.5vLLM: Phi-4 multimodal audio encoder materializes and retains full attention matrices: a single max-length audio OOM-crashes the vLLM engine
CVE-2026-115917Low· 3.4vLLM: Permissive default CORS origins allow cross-origin requests to the vLLM HTTP server
CVE-2026-115915High· 8.1vLLM: Untrusted EPS inputs can trigger denial of service or Ghostscript RCE
CVE-2026-115914Medium· 6.5vLLM DeepSeek-OCR image processing vulnerable to denial of service