UTC

vLLM

By vLLM. 164 published vulnerabilities.

Advisories
164
Critical
9
Exploited in the wild
None listed
Latest
2026-11-18
Packages it ships
Vulnerabilities
SeverityVulnerabilityPublishedFixed in
Medium/v1/systemone bypasses the --trust-request-chat-template gateGHSA-xgh8-qwjh-q8532026-11-180.34.0
MediumThinking-budget sparse state aliases onto a co-batched request during a mixed `SWAP`, forcing the reasoning-end token into another request'sGHSA-fpch-3xrf-7q962026-11-090.28.0
MediumCross-user token data leakage in async scheduling + Pipeline ParallelismGHSA-v97p-wpxp-px842026-11-090.21.0
MediumResponse amplification in OpenAI validation error handlerGHSA-mwgq-9582-w2vf2026-11-070.30.0
HighMissing request-wide media budget allows single-request frontend resource exhaustionGHSA-5r3c-5x4v-8gjh2026-11-020.33.0
HighUnauthenticated EngineCore crash via out-of-range prompt token id under skip_tokenizer_init (incomplete fix of GHSA-25q3-v2hm-8vpf)GHSA-q9v9-qpjp-fc242026-11-020.33.0
MediumvLLM Speculators Config verifier.name_or_path Element Directory Traversal Information Disclosure VulnerabilityGHSA-qrj2-j723-m6w82026-11-020.33.0
MediumvLLM: DoS via unbounded chat_template Jinja2 rendering exhausts the renderer thread poolCVE-2026-115911 · GHSA-542m-9234-8gvf2026-10-280.32.0
HighvLLM: Public multimodal SHM handle forgery reaches worker-side `pickle.loads()` with insecure serialization disabledCVE-2026-115912 · GHSA-6757-hggm-hp2w2026-10-280.32.0
MediumvLLM: Phi-4 multimodal audio encoder materializes and retains full attention matrices: a single max-length audio OOM-crashes the vLLM engineCVE-2026-115913 · GHSA-8397-8qf6-f2gm2026-10-280.32.0
MediumvLLM DeepSeek-OCR image processing vulnerable to denial of serviceCVE-2026-115914 · GHSA-gq5q-89c7-vhvx2026-10-280.32.0
HighvLLM: Untrusted EPS inputs can trigger denial of service or Ghostscript RCECVE-2026-115915 · GHSA-h3rc-j82p-2c3h2026-10-280.32.0
MediumvLLM: Out-of-vocabulary prompt token id passes input validation and triggers a CUDA device-side assert (engine crash / DoS)CVE-2026-115916 · GHSA-q97c-p368-75c42026-10-280.32.0
LowvLLM: Permissive default CORS origins allow cross-origin requests to the vLLM HTTP serverCVE-2026-115917 · GHSA-vvpg-rrww-wpmh2026-10-280.32.0
MediumvLLM: Responses API bypasses trust_request_chat_template for chat_template_kwargsCVE-2026-115918 · GHSA-x567-49g5-p4cf2026-10-280.32.0
MediumvLLM denial of service via prompt embeds on XD-RoPE models (incomplete fix of CVE-2026-55514)GHSA-5q78-j82c-2vr22026-10-270.30.0
MediumvLLM OpenCVDynamicOpenPanguVideoBackend: request-level num_frames without a ceiling — unauthenticated memory/CPU exhaustion (OOM-kill) on anGHSA-894j-wg8m-63h82026-10-270.31.0
MediumQuadratic Cohere2 Vision preprocessing DoS via request-controlled max_patchesGHSA-9g3c-x747-c6wr2026-10-270.31.0
Mediumvllm p2p never connected zmq_exhaustionCVE-2026-94624 · GHSA-frvg-qw88-hww92026-10-270.32.0
HighvLLM: Request-controlled mm_processor_kwargs.code_revision allows remote code executionCVE-2026-114053 · GHSA-h3rc-6mm3-gc2m2026-10-270.31.0
Mediumsanitize_message() address regex is lowercase-only and does not match PIL's uppercase __repr__, leaking live heap addresses to unauthenticatGHSA-m9fg-w9qh-67pc2026-10-270.31.0
MediumIncomplete CVE-2026-34755 remediation can be bypassed by request-level num_framesGHSA-vxqj-p4gw-9h4c2026-10-270.31.0
MediumMiniMax M3 video preprocessing permits request-driven CPU resource exhaustion through min_pixelsGHSA-w4cg-mhc7-mgw32026-10-270.31.0
HighvLLM from 0.20.1 before 0.29.0 contains an improper input validation vulnerability that allows remote attackers to crash the engine by supplying out-of-vocabulary stoptokenids.CVE-2026-1161812026-10-250.29.0
HighvLLM from 0.8.5 before 0.31.0 has a denial of service flaw.CVE-2026-1161802026-10-250.31.0
HighvLLM 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 numframes through mediaiokwargs.CVE-2026-1161792026-10-250.31.0
HighvLLM 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.CVE-2026-1161782026-10-250.30.0
HighvLLM from 0.10.2 before 0.25.0 contains an uncontrolled resource consumption vulnerability that allows unauthenticated attackers to exhaust memory because the Nemotron-VL processors globally set Image.MAXIMAGEPIXELS to None.CVE-2026-1161772026-10-250.25.0
HighvLLM from 0.9.2 before 0.24.0 contains a server-side request forgery vulnerability that allows API callers to force outbound connections by embedding a decode address in the X-Request-Id header.CVE-2026-1161762026-10-250.24.0
HighvLLM from 0.19.0 before 0.24.0 contains a decompression bomb vulnerability because the AudioMediaIO multimodal path calls loadaudio without the maxdurations decoded-duration limit.CVE-2026-1161752026-10-250.24.0
HighvLLM from 0.22.0 before 0.24.0 contains an improper input validation vulnerability in the Rust gRPC Generate and GenerateStream service that forwards unchecked prompt token IDs to EngineCore.CVE-2026-1161742026-10-250.24.0
HighvLLM from 0.22.0 before 0.24.0 contains a denial of service vulnerability that allows remote attackers to terminate EngineCore by bypassing logprobtokenids limits in the Rust gRPC frontend.CVE-2026-1161732026-10-250.24.0
HighvLLM from 0.22.0 before 0.24.0 contains an improper input validation vulnerability in the Rust gRPC converter that allows remote attackers to bypass maxlogprobs limits via CandidateTokens.topn.CVE-2026-1161722026-10-250.24.0
HighvLLM : DoS via unvalidated mm_processor_kwargs in multimodal image processors (Isaac, InternVL, H2OVL, etc.)CVE-2026-114055 · GHSA-gx7p-2j49-hfq42026-10-230.31.0
MediumvLLM: Forged multimodal UUID + P0/P1 processor-cache eviction drift: unauthenticated remote crash of the whole vLLM engineCVE-2026-114050 · GHSA-p92p-rxj5-7p2x2026-10-230.31.0
HighvLLM 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 cachesalt.CVE-2026-1140542026-10-210.31.0
HighvLLM before 0.31.0 skips the mintokens <= maxtokens check for /inference/v1/generate requests with a null maxtokens, so the EngineCore input thread later dies on an uncaught VLLMValidationError while /health still returns 200.CVE-2026-1140512026-10-210.31.0
MediumvLLM 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 allowedtokenids rows.CVE-2026-1140492026-10-210.31.0
MediumRemote DoS in disaggregated multimodal /inference/v1/generate via malicious kwargs_data / mm_placeholdersGHSA-34qw-q2h2-jp3m2026-10-090.30.0
HighFrame-count cap bypass in VideoMediaIO.load_base64 for video/jpeg via media_io_kwargs.video.num_framesGHSA-4m6g-h9mv-wqjc2026-10-090.31.0
MediumRust gRPC output_candidates.token_ids bypasses logprob_token_ids limits and can terminate EngineCoreGHSA-6pgw-f3pv-4h8q2026-10-090.24.0
HighRemote DoS in vLLM via allowed_token_ids Speculative-Decoding Metadata MismatchGHSA-8c65-hq7q-r7jm2026-10-090.18.0
HighNemotron-VL processors disable Pillow decompression-bomb guard globally, enabling unauthenticated DoSGHSA-mm38-5g96-7j6x2026-10-090.25.0
MediumMultimodal chat audio decoding bypasses the decompression-bomb duration limitGHSA-r45p-7qwx-qwjw2026-10-090.24.0
HighCross-tenant DoS via uninitialized LogprobsTensors with prompt_logprobsGHSA-rhcx-5729-88vg2026-10-090.31.0
MediumRust gRPC CandidateTokens.top_n bypasses max_logprobs and can terminate EngineCoreGHSA-rwfw-xvg3-39372026-10-090.24.0
HighSSRF via X-Request-Id header drives P2pNcclConnector outbound ZMQ+NCCL pair, leaking KV-cache tensorsGHSA-vfp2-c8pq-v6h62026-10-090.24.0
MediumRust gRPC Generate accepts out-of-vocabulary prompt token IDs and can terminate EngineCoreGHSA-w2v2-5662-8v5q2026-10-090.24.0
MediumUnvalidated stop_token_ids causes engine crash (DoS) and out-of-bounds GPU writeGHSA-w9r6-29q4-pvpr2026-10-090.29.0
LowA security flaw has been discovered in vllm-project vLLM up to 0.31.0.CVE-2026-1059222026-10-06See advisory
MediumAn oversized min_tokens in one request hangs the vLLM engine while /health stays greenGHSA-4xqp-c3mv-qff72026-10-060.31.0
LowV1 InputBatch.condense leaves a stale allowed_token_ids mask on recycled batch rowsGHSA-6cxc-2vcg-w5qc2026-10-060.31.0
HighMOSS-Audio request-controlled processor cache permits remote memory exhaustionCVE-2026-114052 · GHSA-823j-m4cj-hmjf2026-10-060.31.0
LowvLLM: Harmony tool continuations drop `cache_salt` — restoring a cross-tenant prefix-cache membership oracleCVE-2026-105752 · GHSA-935w-9g4m-p28p2026-10-060.30.0
MediumvLLM: Mirrored multimodal IPC caches desync after a rejected request — a later request reusing the same media hash trips a receiver assertioCVE-2026-105753 · GHSA-ph3r-5jfg-f84f2026-10-060.28.0
MediumCaller-controllable values become or collide with shared cache keys, enabling cross-request cache aliasing (2 sites: multimodal EXIF hash anGHSA-rh6f-3x46-j33q2026-10-060.31.0
MediumvLLM is an inference and serving engine for large language models.CVE-2026-1057592026-10-050.30.0
MediumvLLM: Loose `cache_salt` validation lets a single request kill EngineCore on LMCache-MP deployments — uncaught downstream `ValueError` deniaCVE-2026-105756 · GHSA-2823-qmq8-rwvj2026-10-050.30.0
MediumvLLM: Flash late-interaction scoring caches query embeddings under a caller-controlled request id — cross-request integrity break and induceCVE-2026-105755 · GHSA-2phq-3phc-84px2026-10-050.30.0
MediumvLLM: GLMGA video sampling permits request-driven CPU and memory exhaustionCVE-2026-105760 · GHSA-58v5-2m8f-94pr2026-10-050.30.0
MediumvLLM: Structured-output request errors escape the request boundary and terminate the shared EngineCore — engine-fatal denial of service (3 sCVE-2026-105757 · GHSA-85xf-c7hm-whqw2026-10-050.30.0
MediumvLLM: Scale-out disaggregated multimodal transport trusts caller-supplied featuresCVE-2026-105754 · GHSA-ph72-cqr5-qpp72026-10-050.30.0
MediumvLLM: Qwen2-VL / Qwen3-VL video samplers bound on request-controlled max_frames, which the num_frames ceiling does not reachCVE-2026-105758 · GHSA-x6mc-67gf-chw42026-10-050.30.0
MediumA flaw has been found in vllm-project vLLM up to 0.26.0.CVE-2026-1032412026-09-30See advisory
MediumCrafted IAMF audio upload reaches a PyAV/FFmpeg native heap overflow through the speech transcription path — denial of serviceGHSA-m52c-39rh-f3gp2026-09-280.29.0
HighvLLM before 0.29.0 accepts user-controlled stoptokenids on the OpenAI-compatible POST /v1/completions and POST /v1/chat/completions endpoints but validates only that the values are integers, not that each token id is within the model…CVE-2026-1006542026-09-260.29.0
HighvLLM is an inference and serving engine for large language models.CVE-2026-1006532026-09-260.28.0
HighvLLM versions 0.22.0 through 0.23.0 fail to validate stoptokenids against vocabulary bounds in Rust HTTP and gRPC frontends, allowing out-of-vocabulary token IDs to reach MinTokensLogitsProcessor.CVE-2026-1006522026-09-260.24.0
HighvLLM before 0.29.0 fails to enforce decoder prompt-length validation on the disaggregated serving endpoint /inference/v1/generate.CVE-2026-1006512026-09-260.29.0
HighvLLM through 0.29.0 fetches and fully materializes remote or inline media before enforcing its documented media controls (the VLLMMAXAUDIOCLIPFILESIZEMB compressed-audio size cap, default 25 MB, and the per-modality --limit-mm-per-prompt…CVE-2026-1006502026-09-26See advisory
HighvLLM Mooncake connector through 0.29.0 fails to properly manage GPU KV cache block ownership when concurrent child requests share a single transfer ID in prefill/decode disaggregated deployments.CVE-2026-946272026-09-21See advisory
HighvLLM through 0.29.0 fails to validate the tpsize parameter in kvtransferparams on OpenAI-compatible completion endpoints, allowing attackers to allocate unbounded memory.CVE-2026-946262026-09-21See advisory
MediumvLLM through 0.29.0 contains a resource exhaustion vulnerability in MooncakeConnector where rejected prefill requests create ownerless transfer placeholders that are never reclaimed.CVE-2026-946252026-09-21See advisory
HighvLLM through 0.29.0 contains a denial of service vulnerability in the NIXL connector's prefix caching implementation that fails to properly validate block counts across multi-prompt completion requests in prefill/decode disaggregated…CVE-2026-946232026-09-21See advisory
HighvLLM versions through 0.29.0 contain a denial of service vulnerability in the NIXL connector's metadata handling for prefill/decode disaggregated deployments.CVE-2026-946222026-09-21See advisory
LowvLLM through 0.29.0 fails to properly validate badwords token indices against the model's generation output width in SamplingParams.updatefrom_tokenizer().CVE-2026-939892026-09-19See advisory
MediumvLLM through 0.29.0 contains a memory corruption vulnerability in the Triton bincountkernel where prompt token IDs index the penalty prompt-presence bitset without bounds checking against vocabulary size.CVE-2026-938412026-09-18See advisory
MediumvLLM before 0.29.0 validates allowedtokenids against tokenizer length instead of model output logits width in SamplingParams.validateallowedtokenids().CVE-2026-938402026-09-180.29.0
HighvLLM through 0.29.0 fails to properly clean up decode-side metadata for rejected inference requests in prefill/decode disaggregated deployments.CVE-2026-934362026-09-17See advisory
MediumvLLM: Request-selected PyNvVideoCodec GPU decode bypasses static VRAM reservationCVE-2026-69147 · GHSA-8pw2-6jv3-mj5j2026-09-170.28.0
MediumvLLM: Unauthenticated audio decompression-bomb DoS in /v1/chat/completionsCVE-2026-57173 · GHSA-hcwq-8wjf-3gcr2026-09-160.24.0
MediumDisaggregated generate skips decoder prompt-length validation for some multimodal processorsGHSA-3mqx-f33v-vgp92026-09-120.29.0
MediumIncomplete artifact pin propagation in FunAudioChat and Tarsier2GHSA-hhv2-872h-628q2026-09-120.28.0
LowSampler Subclass Counter Shadowing Bypasses PyNvVideoCodec Decoder Limits and GPU Memory AccountingCVE-2026-100649 · GHSA-j682-9xp5-rrf32026-09-120.29.0
MediumUncontrolled resource consumption: multimodal chat audio decoding ignores `VLLM_MAX_AUDIO_CLIP_FILESIZE_MB`CVE-2026-100648 · GHSA-jcq2-4gch-5qhf2026-09-120.29.0
MediumRemote media is fetched and fully materialized before the documented media size / item limits are enforced — remote pre-inference memory andGHSA-p6g9-7v3x-m8mv2026-09-12See advisory
MediumRust HTTP/gRPC stop_token_ids bypass Python vocab-bound fix and can terminate EngineCoreGHSA-qff2-492f-9fm42026-09-120.24.0
MediumOut-of-range `stop_token_ids` with `min_tokens` can kill vLLM EngineCoreGHSA-v5gm-qgmv-gc6c2026-09-120.29.0
MediumvLLM DoS via unbounded `cache_salt` length: multi-hundred-MB salt stalls the single EngineCore scheduler thread (CPU exhaustion)CVE-2026-100647 · GHSA-wpww-v874-ph2p2026-09-120.29.0
MediumvLLM: SSRF + arbitrary local file read in MiMoV2OmniMultiModalProcessor `_fetch_image` and audio loader bypass MediaConnector protectionsCVE-2026-73560 · GHSA-4hhp-h66f-j5j72026-09-080.26.0
MediumvLLM: Cross-User Data Leak VulnerabilityCVE-2026-73558 · GHSA-7m6h-x95x-82q52026-09-080.27.0
MediumvLLM: ReDoS via structured_outputs.regex in the lm-format-enforcer backend (no compile timeout) — missed sibling of GHSA-rwxx-mrjm-wc2mCVE-2026-73556 · GHSA-48jh-3gj7-fg8v2026-09-040.26.0
MediumvLLM: Derender endpoints decode caller-supplied GenerateResponse token IDs without output boundsCVE-2026-71486 · GHSA-8737-qx52-hjff2026-09-040.26.0
MediumvLLM: Unauthenticated Internal Path and Username Disclosure via Validation Error MessagesCVE-2026-73555 · GHSA-hwrm-c4cx-rf4j2026-09-040.26.0
MediumvLLM: Incomplete CVE-2025-62164 remediation can be bypassed by concurrent prompt partsCVE-2026-73557 · GHSA-pr7f-p5mw-fc872026-09-040.26.0
HighDenial of service: a negative token id in `/v1/embeddings` or `/pooling` input kills the vLLM engine via a CUDA device-side assertionCVE-2026-93592 · GHSA-25q3-v2hm-8vpf2026-09-030.28.0
HighvLLM up to and including 0.17.0 allows remote attackers to cause a Denial of Service via memory exhaustion.CVE-2026-372372026-08-28See advisory
HighLlavaOnevision2 processor loader executes attacker model code with `trust_remote_code=False` (inert `trust_remote_code` kwarg to `transformeCVE-2026-90553 · GHSA-3c86-2m5g-59q72026-08-280.28.0
MediumDenial of Service NanoNemoTronVL Video Audio Extraction BombCVE-2026-90554 · GHSA-936p-m5pv-vvjf2026-08-280.28.0
MediumSpeech-to-text audio decode duration limit bypass via forged header sample rateCVE-2026-90555 · GHSA-99f2-hwrc-gvq82026-08-280.28.0
MediumvLLM: Completion prompt lists fan out into unbounded engine requestsCVE-2026-73559 · GHSA-87x5-vmc3-756j2026-08-130.26.0
MediumvLLM Unauthenticated Requests Exploit DeepStream Backend Confusion for DoSCVE-2026-78684 · GHSA-cqm8-jxg6-fqfq2026-08-110.27.0
HighvLLM denial of service via prompt embeds on M-RoPE modelsCVE-2026-55514 · GHSA-33cg-gxv8-3p8g2026-07-200.24.0
MediumvLLM: Processing differential in multi-channel audio downmixing enables hidden-input/moderation bypass for audio modelsCVE-2026-34760 · GHSA-6c4r-fmh3-7rh82026-07-170.18.0
HighvLLM has Remote DoS via Invalid Recovered Token ReinjectionCVE-2026-54234 · GHSA-8wr5-jm2h-8r4f2026-07-170.24.0
HighvLLM: ReDoS via structured_outputs.regex compiled without timeout in xgrammar and outlines backendsCVE-2026-55574 · GHSA-rwxx-mrjm-wc2m2026-07-170.24.0
MediumvLLM: Speech-to-text upload size limit is enforced after full UploadFile readCVE-2026-55646 · GHSA-v82g-2437-67m22026-07-170.24.0
HighvLLM is an inference and serving engine for large language models (LLMs).CVE-2026-542322026-06-220.22.1
MediumvLLM versions >= 0.6.3 and < 0.9.0 contain multiple regular expression denial of service (ReDoS) vulnerabilities.CVE-2025-713792026-06-200.9.0
MediumvLLM: GGUF dequantize kernel int truncation exposes uninitialized GPU memory in multi-tenant servingCVE-2026-53923 · GHSA-5jv2-g5wq-cmr42026-06-170.24.0
MediumvLLM: OOM Denial of Service via Audio Decompression BombCVE-2026-54233 · GHSA-6pr9-rp53-2pmc2026-06-170.24.0
MediumvLLM: temperature=NaN and temperature=Infinity bypass validation and propagate to GPU kernelsCVE-2026-54235 · GHSA-7h4p-rffg-78232026-06-170.24.0
MediumvLLM: image EXIF Rotation & PNG tRNS Transparency Not Normalized, Causing Mismatch Between Model Input and ExpectationsCVE-2026-12491 · GHSA-8jr5-v98p-w75m2026-06-170.24.0
MediumvLLM: incomplete CVE-2026-22778 fix leaks PIL repr addresses via Anthropic routerCVE-2026-54236 · GHSA-hgg8-fqqc-vfmw2026-06-170.24.0
CriticalvLLM: OpenAI auth bypassCVE-2026-48746 · GHSA-94f4-hr76-p5j62026-06-160.22.0
HighvLLM: Security Check Bypass via assert Statement in Activation Function Loading Allows Arbitrary Code ExecutionCVE-2026-41523 · GHSA-q8gq-377p-jq3r2026-06-160.22.0
HighvLLM is vulnerable to an Out-of-Memory (OOM) Denial of Service (DoS) attack due to unbounded frame count processing in the `VideoMediaIO.loaCVE-2026-5497 · GHSA-wcwg-c5fc-9vrc2026-06-110.19.0
MediumvLLM's Artifact Pin Decay allows pinned deployments to load unpinned code, weights, and processorsCVE-2026-47155 · GHSA-3ww4-5jv9-j5gm2026-06-100.22.0
Mediumvllm has Improper Resource Shutdown or ReleaseCVE-2026-9540 · GHSA-98f3-hwg4-4rf72026-05-26See advisory
MediumvLLM: extract_hidden_states speculative decoding crashes server on any request with penalty parametersCVE-2026-44223 · GHSA-83vm-p52w-f9pw2026-05-060.20.0
MediumvLLM Vulnerable to Remote DoS via Special-Token PlaceholdersCVE-2026-44222 · GHSA-hpv8-x276-m59f2026-05-050.20.0
LowvLLM makes Use of Uninitialized ResourceCVE-2026-7141 · GHSA-x368-4g9h-fvv42026-04-270.19.1
MediumvLLM: Unauthenticated OOM Denial of Service via Unbounded `n` Parameter in OpenAI API ServerCVE-2026-34756 · GHSA-3mwp-wvh9-75282026-04-030.19.0
MediumvLLM: Server-Side Request Forgery (SSRF) in `download_bytes_from_url `CVE-2026-34753 · GHSA-pf3h-qjgv-vcpr2026-04-030.19.0
MediumvLLM: Denial of Service via Unbounded Frame Count in video/jpeg Base64 ProcessingCVE-2026-34755 · GHSA-pq5c-rjhq-qp7p2026-04-030.19.0
HighvLLM has Hardcoded Trust Override in Model Files Enables RCE Despite Explicit User Opt-OutCVE-2026-27893 · GHSA-7972-pg2x-xr592026-03-270.18.0
MediumvLLM has SSRF Protection BypassCVE-2026-25960 · GHSA-v359-jj2v-j5362026-03-090.17.0
CriticalvLLM has RCE In Video ProcessingCVE-2026-22778 · GHSA-4r2x-xpjr-7cvv2026-02-020.14.1
HighvLLM vulnerable to Server-Side Request Forgery (SSRF) through MediaConnectorCVE-2026-24779 · GHSA-qh4c-xf7m-gxfc2026-01-280.14.1
HighvLLM affected by RCE via auto_map dynamic module loading during model initializationCVE-2026-22807 · GHSA-2pc9-4j83-qjmr2026-01-210.14.0
MediumvLLM is vulnerable to DoS in Idefics3 vision models via image payload with ambiguous dimensionsCVE-2026-22773 · GHSA-grg2-63fw-f2qr2026-01-130.12.0
HighvLLM introduced enhanced protection for CVE-2025-62164CVE-2026-56340 · GHSA-mcmc-2m55-j8jj2026-01-080.13.0
HighvLLM vulnerable to remote code execution via transformers_utils/get_configCVE-2025-66448 · GHSA-8fr4-5q9j-m8gm2025-12-020.11.1
MediumvLLM vulnerable to DoS via large Chat Completion or Tokenization requests with specially crafted `chat_template_kwargs`CVE-2025-62426 · GHSA-69j4-grxj-j64p2025-11-200.11.1
HighvLLM deserialization vulnerability leading to DoS and potential RCECVE-2025-62164 · GHSA-mrw7-hf4f-83pf2025-11-200.11.1
HighvLLM vulnerable to DoS with incorrect shape of multimodal embedding inputsCVE-2025-62372 · GHSA-pmqf-x6x8-p7qw2025-11-200.11.1
HighvLLM is vulnerable to Server-Side Request Forgery (SSRF) through `MediaConnector` classCVE-2025-6242 · GHSA-3f6c-7fw2-ppm42025-10-070.11.0
MediumvLLM: Resource-Exhaustion (DoS) through Malicious Jinja Template in OpenAI-Compatible ServerCVE-2025-61620 · GHSA-6fvq-23cw-56282025-10-070.11.0
HighvLLM is vulnerable to timing attack at bearer authCVE-2025-59425 · GHSA-wr9h-g72x-mwhm2025-10-070.11.0
HighvLLM has remote code execution vulnerability in the tool call parser for Qwen3-CoderCVE-2025-9141 · GHSA-79j6-g2m3-jgfw2025-08-210.10.1.1
Highvllm API endpoints vulnerable to Denial of Service AttacksCVE-2025-48956 · GHSA-rxc4-3w6r-4v472025-08-210.10.1.1
LowPotential Timing Side-Channel Vulnerability in vLLM’s Chunk-Based Prefix CachingCVE-2025-46570 · GHSA-4qjh-9fv9-r85r2025-05-280.9.0
MediumvLLM DOS: Remotely kill vllm over http with invalid JSON schemaCVE-2025-48942 · GHSA-6qc9-v4r8-22xg2025-05-280.9.0
MediumvLLM allows clients to crash the openai server with invalid regexCVE-2025-48943 · GHSA-9hcf-v7m4-6m2j2025-05-280.9.0
MediumvLLM has a Weakness in MultiModalHasher Image Hashing ImplementationCVE-2025-46722 · GHSA-c65p-x677-fgj62025-05-280.9.0
MediumvLLM vulnerable to Regular Expression Denial of ServiceGHSA-j828-28rj-hfhp2025-05-280.9.0
MediumvLLM Tool Schema allows DoS via Malformed pattern and type FieldsCVE-2025-48944 · GHSA-vrq3-r879-7m652025-05-280.9.0
MediumvLLM has a Regular Expression Denial of Service (ReDoS, Exponential Complexity) Vulnerability in `pythonic_tool_parser.py`CVE-2025-48887 · GHSA-w6q7-j642-7c252025-05-280.9.0
CriticalvLLM Allows Remote Code Execution via PyNcclPipe Communication ServiceCVE-2025-47277 · GHSA-hjq4-87xh-g4fv2025-05-200.8.5
HighRemote Code Execution Vulnerability in vLLM Multi-Node Cluster ConfigurationCVE-2025-30165 · GHSA-9pcc-gvx5-r5wm2025-05-060.10.0
HighData exposure via ZeroMQ on multi-node vLLM deploymentCVE-2025-30202 · GHSA-9f8f-2vmf-885j2025-04-290.8.5
CriticalvLLM Vulnerable to Remote Code Execution via Mooncake IntegrationCVE-2025-32444 · GHSA-hj4w-hm2g-p6w52025-04-290.8.5
MediumvLLM: Quadratic Time Complexity in Input Token Processing​ leads to denial of serviceCVE-2025-46560 · GHSA-vc6m-hm49-g9qg2025-04-290.8.5
CriticalCVE-2025-24357 Malicious model remote code execution fix bypass with PyTorch < 2.6.0GHSA-ggpf-24jw-3fcw2025-04-230.8.0
MediumvLLM vulnerable to Denial of Service by abusing xgrammar cacheGHSA-hf3c-wxg2-49q92025-04-150.8.4
CriticalvLLM Deserialization of Untrusted Data vulnerabilityCVE-2024-11041 · GHSA-5vqr-wprc-cpp72025-03-20See advisory
CriticalvLLM allows Remote Code Execution by Pickle Deserialization via AsyncEngineRPCServer() RPC server entrypointsCVE-2024-9053 · GHSA-cj47-qj6g-x7r42025-03-20See advisory
CriticalvLLM deserialization vulnerability in vllm.distributed.GroupCoordinator.recv_objectCVE-2024-9052 · GHSA-pgr7-mhp5-fgjp2025-03-20See advisory
MediumvLLM denial of service via outlines unbounded cache on diskCVE-2025-29770 · GHSA-mgrm-fgjv-mhv82025-03-190.8.0
CriticalvLLM Allows Remote Code Execution via Mooncake IntegrationCVE-2025-29783 · GHSA-x3m8-f7g5-qhm72025-03-190.8.0
LowvLLM uses Python 3.12 built-in hash() which leads to predictable hash collisions in prefix cacheCVE-2025-25183 · GHSA-rm76-4mrf-v9r82025-02-060.7.2
Highvllm: Malicious model to RCE by torch.load in hf_model_weights_iteratorCVE-2025-24357 · GHSA-rh4j-5rhw-hr542025-01-270.7.0
HighvLLM denial of service vulnerabilityCVE-2024-8768 · GHSA-w2r7-9579-27hf2024-09-170.5.5
MediumvLLM Denial of Service via the best_of parameterCVE-2024-8939 · GHSA-wc36-9694-f9rf2024-09-17See advisory

CVEs matched by the NVD's product names, checked 2026-11-18. Where the vendor runs the service, its fix reaches you without an update.

Sources: GitHub Advisory Database, reviewed and malware advisories (CC BY 4.0, https://github.com/advisories); advisories published by Cursor (https://github.com/cursor/cursor/security/advisories); advisories published by OpenClaw (https://github.com/openclaw/openclaw/security/advisories); advisories published by vLLM (https://github.com/vllm-project/vllm/security/advisories); advisories published by Open WebUI (https://github.com/open-webui/open-webui/security/advisories); advisories published by n8n (https://github.com/n8n-io/n8n/security/advisories); advisories published by Flowise (https://github.com/FlowiseAI/Flowise/security/advisories); advisories published by Langflow (https://github.com/langflow-ai/langflow/security/advisories); advisories published by the MCP Python SDK project (https://github.com/modelcontextprotocol/python-sdk/security/advisories); advisories published by MLflow (https://github.com/mlflow/mlflow/security/advisories); CISA Known Exploited Vulnerabilities catalog (CC0). Data as of 2026-11-22. ShieldMap describes how each problem works and how to fix it; it does not publish exploit code. Report a correction to [email protected].

This product uses the NVD API but is not endorsed or certified by the NVD. Product facts from github.com, checked 2026-11-18. Logos are their owners' marks, shown only to identify the products.