vllm-project has 23 CVEs on record. Disclosure cadence is accelerating: 23 in the last 90 days against 0 in the 90 before. The busiest recent month was October 2026 with 18. The median CVSS is 6.5 (medium). None have a confirmed exploitation report. The dominant weakness classes are CWE-770 (8) and CWE-400 (5).
CVEs per month
Last 12 months, by publish date
- Exploited share
- 0% vs 1% corpus
- Median CVSS
- 6.5
- Publish → KEV
- —
- Last 90 days
- 23 prev 0
Worst active — by depth score
CVE-2026-109311Medium· 6.5vLLM before 0.29.0 passes uploaded audio to the PyAV/FFmpeg av.open() fallback with only a file-size check, exposing FFmpeg's IAMF parser heap overflow48CVE-2026-100653Medium· 6.5vLLM is an inference and serving engine for large language models48CVE-2026-115915High· 8.1vLLM: Untrusted EPS inputs can trigger denial of service or Ghostscript RCE45CVE-2026-116180High· 7.7vLLM from 0.8.5 before 0.31.0 Cross-Tenant DoS via Uninitialized LogprobsTensors42CVE-2026-115912High· 7.5vLLM: Public multimodal SHM handle forgery reaches worker-side `pickle.loads()` with insecure serialization disabled41
vllm-project vulnerabilities
CVEs affecting vllm-project, newest first. Open any entry for full detail, references, and exploit status.
23 CVEsRSS
CVE-2026-115911Medium· 6.5vLLM: DoS via unbounded chat_template Jinja2 rendering exhausts the renderer thread pool
vLLM is an inference and serving engine for large language models. From 0.19.0 until 0.32.0, authenticated callers can supply a request chat template when no server-side --chat-template is configured, or a malicious model repository can …
CVE-2026-115912High· 7.5vLLM: Public multimodal SHM handle forgery reaches worker-side `pickle.loads()` with insecure serialization disabled
vLLM is an inference and serving engine for large language models. Prior to 0.32.0, an authenticated caller can place forged shared-memory address and monotonic_id values in public multimodal features.kwargs_data. In scale-out deployment…
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
vLLM is an inference and serving engine for large language models. Prior to 0.32.0, the Phi-4 multimodal audio encoder in phi4mm_audio.py and phi4mm_utils.py uses eager attention by default, allowing a remote client to submit one legal l…
CVE-2026-115916Medium· 6.5vLLM: Out-of-vocabulary prompt token id passes input validation and triggers a CUDA device-side assert (engine crash / DoS)
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 valid…
CVE-2026-115917Low· 3.4vLLM: Permissive default CORS origins allow cross-origin requests to the vLLM HTTP server
vLLM is an inference and serving engine for large language models. Prior to 0.32.0, the OpenAI-compatible server configures Starlette CORSMiddleware with allowed_origins set to a wildcard by default. A malicious web page can use a victim…
CVE-2026-115915High· 8.1vLLM: Untrusted EPS inputs can trigger denial of service or Ghostscript RCE
vLLM is an inference and serving engine for large language models. Prior to 0.32.0, ImageMediaIO passes attacker-controlled image bytes to Pillow through Image.open(BytesIO(data)) without restricting decoder selection, and the MIME label…
CVE-2026-115914Medium· 6.5vLLM DeepSeek-OCR image processing vulnerable to denial of service
vLLM is an inference and serving engine for large language models. From 0.11.1 until 0.32.0, the DeepSeek-OCR, DeepSeek-OCR-2, and Unlimited-OCR image path allows a remote client that can reach the OpenAI-compatible API to submit a legal…
CVE-2026-115918Medium· 6.5vLLM: Responses API bypasses trust_request_chat_template for chat_template_kwargs
vLLM is an inference and serving engine for large language models. From 0.16.0 until 0.32.0, the /v1/responses endpoint allows an authenticated caller to place an attacker-controlled template in chat_template_kwargs[chat_template] when n…
CVE-2026-116175Medium· 6.5vLLM 0.19.0 before 0.24.0 Audio Decompression Bomb DoS via Chat Completions
vLLM from 0.19.0 before 0.24.0 contains a decompression bomb vulnerability because the AudioMediaIO multimodal path calls load_audio without the max_duration_s decoded-duration limit. Authenticated attackers can send small compressed aud…
CVE-2026-116173Medium· 5.9vLLM 0.22.0 before 0.24.0 Denial of Service via gRPC output_candidates.token_ids
vLLM from 0.22.0 before 0.24.0 contains a denial of service vulnerability that allows remote attackers to terminate EngineCore by bypassing logprob_token_ids limits in the Rust gRPC frontend. Attackers can send Generate or GenerateStream…
CVE-2026-116179High· 7.5vLLM 0.18.0 before 0.31.0 Memory Exhaustion via media_io_kwargs num_frames Override
vLLM from 0.18.0 before 0.31.0 contains a resource exhaustion vulnerability that allows API clients to bypass the video/jpeg frame-count cap by overriding num_frames through media_io_kwargs. Attackers can set num_frames to -1 with a larg…
CVE-2026-116178Medium· 5.9vLLM 0.20.0 before 0.30.0 DoS via /inference/v1/generate mm_placeholders
vLLM from 0.20.0 before 0.30.0 contains a denial of service vulnerability that allows remote attackers to crash the engine by sending inconsistent multimodal metadata to the disaggregated /inference/v1/generate endpoint. Attackers can re…
CVE-2026-116180High· 7.7vLLM from 0.8.5 before 0.31.0 Cross-Tenant DoS via Uninitialized LogprobsTensors
vLLM from 0.8.5 before 0.31.0 has a denial of service flaw. LogprobsTensors.empty_cpu() in vllm/v1/outputs.py allocates an uninitialized int32 tensor, and the V1 model runner leaves prompt positions reported as computed by a KV connector…
CVE-2026-114049Low· 3.7vLLM before 0.31.0 Stale allowed_token_ids Mask via InputBatch.condense
vLLM before 0.31.0 contains a cross-request mask leak in InputBatch.condense() that allows remote attackers to constrain other clients' outputs by leaving stale allowed_token_ids rows. Attackers can submit concurrent constrained generate…
CVE-2026-114054Medium· 6.5vLLM before 0.31.0 Cross-Request Cache Aliasing via EXIF ImageID and Prefix-Cache Keys
vLLM before 0.31.0 contains a cache aliasing vulnerability that allows authenticated attackers to collide shared cache keys by supplying caller-controlled values such as EXIF ImageID or cache_salt. Attackers can set a UUID EXIF ImageID t…
CVE-2026-114055High· 7.5vLLM 0.7.3 before 0.31.0 DoS via Unvalidated mm_processor_kwargs
vLLM from 0.7.3 before 0.31.0 contains an uncontrolled resource consumption vulnerability that allows remote clients to control multimodal processor parameters by supplying per-request mm_processor_kwargs. Attackers can set large image s…
CVE-2026-109311Medium· 6.5PoCvLLM before 0.29.0 passes uploaded audio to the PyAV/FFmpeg av.open() fallback with only a file-size check, exposing FFmpeg's IAMF parser heap overflow
vLLM before 0.29.0 passes uploaded audio to the PyAV/FFmpeg av.open() fallback with only a file-size check, exposing FFmpeg's IAMF parser heap overflow. Authenticated remote attackers can repeatedly upload malformed IAMF containers to /v…
CVE-2026-105775Medium· 4.3PoCA security vulnerability has been detected in vllm-project vLLM up to 0.31.0
A security vulnerability has been detected in vllm-project vLLM up to 0.31.0. This impacts the function conv_ssm_forward of the file vllm/model_executor/layers/mamba/mamba_mixer2.py of the component Completions Request Handler. The manip…
CVE-2026-100653Medium· 6.5PoCvLLM is an inference and serving engine for large language models
vLLM is an inference and serving engine for large language models. In versions from 0.22.1 through 0.28.0, the operator-supplied model revision pin (--revision / --code-revision) is not propagated to several Hugging Face artifact loads f…
CVE-2026-92365Medium· 4.3⚖ disputedA vulnerability was found in vllm-project vllm up to 0.29.0
A vulnerability was found in vllm-project vllm up to 0.29.0. Affected by this issue is some unknown functionality of the file vllm/v1/sample/thinking_budget_state.py. The manipulation results in inefficient algorithmic complexity. It is …
CVE-2026-92220Medium· 5.3⚖ disputedA vulnerability was found in vllm-project vLLM 0.26.0/0.27.0
A vulnerability was found in vllm-project vLLM 0.26.0/0.27.0. Affected is the function MoRIIOConnectorScheduler.request_finished/MoRIIOConnectorWorker.get_finished/MoRIIOWrapper._handle_release_message of the file vllm/distributed/kv_tra…
CVE-2026-90878Medium· 4.3PoCA vulnerability was determined in vllm-project vLLM up to 0.27.1
A vulnerability was determined in vllm-project vLLM up to 0.27.1. This affects an unknown part of the file /v1/chat/completions of the component Jinja Template Rendering. This manipulation of the argument chat_template causes resource co…
CVE-2026-90713Low· 3.3PoCA security flaw has been discovered in vllm-project vLLM up to 0.29.0
A security flaw has been discovered in vllm-project vLLM up to 0.29.0. The affected element is the function TiktokenTokenizer::new of the file rust/src/text/src/backend/hf/mod.rs of the component tiktoken vocab File Handler. The manipula…