VulnSea

vLLM vulnerabilities

CVEs whose affected-version data names the vLLM package (pip). Each record lists the affected and patched versions; check a specific version with the dependency checker or POST /api/sbom.

82 CVEsRSS

CVE-2026-94627High· 7.5
today

vLLM 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

vLLM 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. Attackers can trigger GPU memory exhausti…

Twilightvllm-project · vllmvia NVD
CVE-2026-94626High· 7.5
today

vLLM through 0.29.0 fails to validate the tp_size parameter in kv_transfer_params on OpenAI-compatible completion endpoints, allowing attackers to allocate unbounded memory

vLLM through 0.29.0 fails to validate the tp_size parameter in kv_transfer_params on OpenAI-compatible completion endpoints, allowing attackers to allocate unbounded memory. Attackers can supply arbitrary tp_size values in prefill/decode…

Twilightvllm-project · vllmvia NVD
CVE-2026-94625Medium· 5.3
today

vLLM through 0.29.0 contains a resource exhaustion vulnerability in MooncakeConnector where rejected prefill requests create ownerless transfer placeholders that are never reclaimed

vLLM through 0.29.0 contains a resource exhaustion vulnerability in MooncakeConnector where rejected prefill requests create ownerless transfer placeholders that are never reclaimed. Attackers can send rejected requests to exhaust sender…

Sunlitvllm-project · vllmvia NVD
CVE-2026-94624High· 7.5
today

vLLM through 0.29.0 contains a denial of service vulnerability in P2P KV offloading when OffloadingConnector is configured with TieringOffloadingSpec and a peer-to-peer secondary tier

vLLM through 0.29.0 contains a denial of service vulnerability in P2P KV offloading when OffloadingConnector is configured with TieringOffloadingSpec and a peer-to-peer secondary tier. Attackers can supply arbitrary remote host and port …

Twilightvllm-project · vllmvia NVD
CVE-2026-94623High· 7.5
today

vLLM 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 dep…

vLLM 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 dep…

Twilightvllm-project · vllmvia NVD
CVE-2026-94622High· 7.5
today

vLLM versions through 0.29.0 contain a denial of service vulnerability in the NIXL connector's metadata handling for prefill/decode disaggregated deployments

vLLM versions through 0.29.0 contain a denial of service vulnerability in the NIXL connector's metadata handling for prefill/decode disaggregated deployments. Attackers can send requests with incomplete kv_transfer_params dictionary entr…

Twilightvllm-project · vllmvia NVD
CVE-2026-93989Low· 3.1
2d ago

vLLM through 0.29.0 fails to properly validate bad_words token indices against the model's generation output width in SamplingParams.update_from_tokenizer()

vLLM through 0.29.0 fails to properly validate bad_words token indices against the model's generation output width in SamplingParams.update_from_tokenizer(). Attackers can supply out-of-bounds token indices that corrupt logits memory of …

Sunlitvllm-project · vllmEPSS 0.20%via NVD
CVE-2026-93841Low· 3.7
3d ago

vLLM through 0.29.0 contains a memory corruption vulnerability in the Triton _bincount_kernel where prompt token IDs index the penalty prompt-presence bitset without bounds checking against vocabulary size

vLLM through 0.29.0 contains a memory corruption vulnerability in the Triton _bincount_kernel where prompt token IDs index the penalty prompt-presence bitset without bounds checking against vocabulary size. Attackers can submit multimoda…

Sunlitvllm-project · vllmEPSS 0.24%via NVD
CVE-2026-93840Low· 3.7
3d ago

vLLM before 0.29.0 validates allowed_token_ids against tokenizer length instead of model output logits width in SamplingParams._validate_allowed_token_ids()

vLLM before 0.29.0 validates allowed_token_ids against tokenizer length instead of model output logits width in SamplingParams._validate_allowed_token_ids(). Attackers can supply token IDs above the output vocabulary that pass validation…

Sunlitvllm-project · vllmEPSS 0.25%via NVD
CVE-2026-93592High· 7.5PoC
3d ago

vLLM versions before 0.28.0 fail to validate the lower bound of token IDs in the /v1/embeddings and /pooling endpoints, allowing unauthenticated attackers to crash the engine by submitting negative token IDs

vLLM versions before 0.28.0 fail to validate the lower bound of token IDs in the /v1/embeddings and /pooling endpoints, allowing unauthenticated attackers to crash the engine by submitting negative token IDs. A single request with a nega…

Midnightvllm-project · vllmEPSS 0.38%via NVD
CVE-2026-93436High· 7.5
4d ago

vLLM through 0.29.0 fails to properly clean up decode-side metadata for rejected inference requests in prefill/decode disaggregated deployments

vLLM through 0.29.0 fails to properly clean up decode-side metadata for rejected inference requests in prefill/decode disaggregated deployments. Remote attackers can submit requests with max_tokens=0 to exhaust decode-worker memory witho…

Twilightvllm-project · vllmEPSS 0.54%via NVD
CVE-2026-69147Medium· 6.5PoC
5d ago

vLLM is an inference and serving engine for large language models

vLLM is an inference and serving engine for large language models. Prior to 0.28.0, request bodies for Chat Completions and Responses can set media_io_kwargs.video.video_backend to pynvvideocodec, and MediaConnector.fetch_video forwards …

Twilightvllm-project · vllmEPSS 0.46%via NVD
CVE-2026-57173Medium· 6.5
5d ago

vLLM is an inference and serving engine for large language models

vLLM is an inference and serving engine for large language models. Prior to 0.24.0, the input_audio handling path for /v1/chat/completions calls AudioMediaIO.load_bytes or AudioMediaIO.load_file without passing VLLM_MAX_AUDIO_DECODE_DURA…

Sunlitvllm-project · vllmEPSS 0.66%via NVD
CVE-2026-92365Medium· 4.3⚖ disputed
5d ago

A 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 …

Sunlitvllm-project · vllmEPSS 0.39%via NVD
CVE-2026-92220Medium· 5.3⚖ disputed
5d ago

A 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…

Sunlitvllm-project · vLLMEPSS 0.52%via NVD
CVE-2026-90878Medium· 4.3PoC
6d ago

A 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…

Twilightvllm-project · vLLMEPSS 0.30%via NVD
CVE-2026-90713Low· 3.3PoC
1w ago

A 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…

Twilightvllm-project · vLLMEPSS 0.15%via NVD
CVE-2026-90555Medium· 6.5
1w ago

vLLM versions before 0.28.0 fail to validate audio sample rate headers in the transcription endpoint, allowing authenticated clients to bypass duration checks

vLLM versions before 0.28.0 fail to validate audio sample rate headers in the transcription endpoint, allowing authenticated clients to bypass duration checks. Attackers can submit forged FLAC headers with inflated sample rates to trigge…

Sunlitvllm · vllmEPSS 0.29%via NVD
CVE-2026-90554Medium· 6.2
1w ago

vLLM versions >=0.10.2 and <0.28.0 do not apply any audio decode-size or duration limit when extracting audio from video input for NanoNemotronVL models

vLLM versions >=0.10.2 and <0.28.0 do not apply any audio decode-size or duration limit when extracting audio from video input for NanoNemotronVL models. In nano_nemotron_vl.py, _extract_audio_from_videos calls load_audio_pyav(BytesIO(vi…

Sunlitvllm-project · vLLMEPSS 0.11%via NVD
CVE-2026-90553High· 7.8
1w ago

vLLM 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

vLLM 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. Attackers can craft a malicious model with arbi…

Twilightvllm · vllmEPSS 0.21%via NVD
CVE-2026-73560Medium· 6.5
1mo ago

vLLM is an inference and serving engine for large language models

vLLM is an inference and serving engine for large language models. Prior to 0.26.0, the MiMoV2OmniMultiModalProcessor in vllm/transformers_utils/processors/mimo_v2_omni.py passes attacker-controlled image and audio strings through _fetch…

Sunlitvllm · vllmEPSS 0.32%via NVD
CVE-2026-71486Medium· 4.3
1mo ago

vLLM is an inference and serving engine for large language models

vLLM is an inference and serving engine for large language models. Prior to 0.26.0, the /v1/completions/derender and /v1/chat/completions/derender endpoints accept caller-supplied GenerateResponse objects whose generate_responses, choice…

Sunlitvllm · vllmEPSS 0.34%via NVD
CVE-2026-73558Medium· 5.3
1mo ago

vLLM is an inference and serving engine for large language models

vLLM is an inference and serving engine for large language models. Prior to 0.27.0, an integer overflow in blockIdx.x * 2 * d in activation_kernels.cu can cause act_and_mul_kernel to consume another batched user's input, allowing a reque…

Sunlitvllm · vllmEPSS 0.33%via NVD
CVE-2026-73555Medium· 5.3
1mo ago

vLLM is an inference and serving engine for large language models

vLLM is an inference and serving engine for large language models. Prior to 0.26.0, the validation_exception_handler in vllm/entrypoints/openai/server_utils.py converts FastAPI RequestValidationError objects with str(exc), and sanitize_m…

Sunlitvllm · vllmEPSS 0.26%via NVD
CVE-2026-73556Medium· 5.3⚖ disputed
1mo ago

vLLM is an inference and serving engine for large language models

vLLM is an inference and serving engine for large language models. Prior to 0.26.0, the structured_outputs.regex parameter in vllm/v1/structured_output/backend_lm_format_enforcer.py is passed to lmformatenforcer.RegexParser without compi…

Sunlitvllm · vllmEPSS 0.32%via NVD
CVE-2026-73557Medium
1mo ago

vLLM is an inference and serving engine for large language models

vLLM is an inference and serving engine for large language models. From 0.20.2rc0 until 0.26.0, safe_load_prompt_embeds in vllm/renderers/embed_utils.py uses torch.sparse.check_sparse_tensor_invariants, whose process-global save, enable,…

Sunlitvllm · vllmEPSS 0.25%via NVD
CVE-2026-73559Medium· 6.5
1mo ago

vLLM is an inference and serving engine for large language models

vLLM is an inference and serving engine for large language models. From 0.19.0 until 0.26.0, the /v1/completions CompletionRequest.prompt field in vllm/entrypoints/openai/completion/protocol.py accepts an unbounded list[str] or list[list…

Sunlitvllm · vllmEPSS 0.47%via NVD
CVE-2026-55514High· 6.5
2mo ago

vLLM denial of service via prompt embeds on M-RoPE models

vLLM denial of service via prompt embeds on M-RoPE models

Twilightvllm · vllmEPSS 0.66%via OSV
CVE-2026-55646Medium· 6.5
2mo ago

vLLM is an inference and serving engine for large language models. From 0.22.0 to 0.23.0, the /v1/audio/transcriptions and /v1/audio/tran…

vLLM is an inference and serving engine for large language models. From 0.22.0 to 0.23.0, the /v1/audio/transcriptions and /v1/audio/translations routes call request.file.read() to fully materialize an uploaded audio file into memory bef…

Sunlitvllm · vllmEPSS 0.52%via OSV
CVE-2026-48746Critical· 9.1PoC
3mo ago

vLLM is an inference and serving engine for large language models (LLMs)

vLLM is an inference and serving engine for large language models (LLMs). From 0.3.0 until 0.22.0, a vulnerability in ASGI web servers and starlette's trust on those web servers enables an authentication bypass of the OpenAI API Authenti…

Abyssalvllm · vllmEPSS 1.2%via NVD
vLLM vulnerabilities (CVEs) · VulnSea