vllm-project Vulnerabilities and Affected Products
Vulnerabilities associated with vllm.
Products
Clear product- vllm52 vulnerabilities
- vllm-project/vllm4 vulnerabilities
| Vulnerability | Title and context | CVSS | EPSS | PoCs | Signals | STIX action |
|---|---|---|---|---|---|---|
CVE-2026-73559MEDIUM | vLLM: Completion prompt lists fan out into unbounded engine requestsvLLM 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[int]], prompt_to_seq() in vllm/renderers/inputs/preprocess.py and OnlineRenderer.preprocess_completion() in vllm/renderers/online_renderer.py expand every element, and vllm/entrypoints/openai/completion/serving.py creates one engine generator and response slot p… CWE-400Aug 13, 2026 | CVSS6.5v3.1 | EPSS- | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-73558MEDIUM | vLLM: Cross-User Data Leak VulnerabilityvLLM 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 request processed in the same inference batch to receive a partial or complete copy of another user's inference result. This issue is fixed in version 0.27.0. CWE-190Aug 13, 2026 | CVSS5.3v3.1 | EPSS- | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-73557MEDIUM | vLLM: Incomplete CVE-2025-62164 remediation can be bypassed by concurrent prompt partsvLLM 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, and restore state can be raced by concurrent prompt_embeds parts submitted to POST /v1/chat/completions through AsyncMultiModalItemTracker.resolve_items, asyncio.gather, and the default executor, allowing an invalid sparse tensor to reach tensor.to_dense despite… CWE-362Aug 13, 2026 | CVSS6.3v4.0 | EPSS- | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-73556MEDIUM | vLLM: ReDoS via structured_outputs.regex in the lm-format-enforcer backend (no compile timeout) — missed sibling of CVE-2026-55574vLLM 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 compile_regex_with_timeout or validation in validate_structured_output_request_lm_format_enforcer, allowing an unauthenticated /v1/completions request against the lm-format-enforcer backend to consume a CPU core and stall the structured-output engine path with a catas… | CVSS5.3v3.1 | EPSS- | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-73555MEDIUM | vLLM: Unauthenticated Internal Path and Username Disclosure via Validation Error MessagesvLLM 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_message in vllm/entrypoints/utils.py does not remove traceback-style file paths, allowing unauthenticated malformed JSON requests to /v1/chat/completions, /v1/completions, /tokenize, and /detokenize to disclose the OS username, home and virtual-environment paths, … CWE-209Aug 13, 2026 | CVSS5.3v3.1 | EPSS- | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-55514HIGH | vLLM denial of service via prompt embeds on M-RoPE modelsvLLM is a library for LLM inference and serving. From 0.12.0 to before 0.24.0, sending a pure prompt embeds payload in a /v1/completions request with a model using M-RoPE causes EngineCore to fail an assertion and fatally crash, shutting down the entire server application. Any remote user who is authorized to make a /v1/completions request can make such a request and induce a crash. This issue is fixed in version 0.24.0. CWE-617Jul 6, 2026 | CVSS7.1v4.0 | EPSS0.37% | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-55574HIGH | vLLM: ReDoS via structured_outputs.regex compiled without timeout in xgrammar and outlines backendsvLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Prior to 0.24.0, the structured_outputs.regex API parameter passes a user-supplied regular expression string directly to the grammar compiler backends with no compilation timeout; in the xgrammar backend the string reaches the regex compiler with no guard, and in the outlines backend the validation step blocks structural issues such as lookarounds and backreferences but performs no complexity analysis, so a pat… CWE-1333Jul 6, 2026 | CVSS8.7v4.0 | EPSS0.324% | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-54234HIGH | vLLM: Remote DoS in vLLM via Invalid Recovered Token ReinjectionvLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Prior to 0.24.0, a frontend-legal multi-request speculative decoding workload can cause the rejection sampler to produce a recovered token equal to the model vocabulary size boundary value, which is then converted to negative one when the engine selects the next live token for a request and is written back into the drafter's input ids; that out-of-vocabulary value is later consumed by the model's embedding and … | CVSS7.5v3.1 | EPSS0.363% | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-55646MEDIUM | vLLM speech-to-text endpoints allocate full upload before enforcing the audio file-size limitvLLM 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 before vLLM checks the documented VLLM_MAX_AUDIO_CLIP_FILESIZE_MB compressed upload size limit (default 25 MB) later in the speech-to-text preprocessing step, so an API caller who can reach those routes can submit an oversized multipart upload and cause vLLM to allo… | CVSS6.5v3.1 | EPSS0.289% | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-47155MEDIUM | vLLM: Artifact Pin Decay in vLLM allows pinned deployments to load unpinned code, weights, and processorsvLLM is an inference and serving engine for large language models (LLMs). Prior to 0.22.0, vLLM's revision pinning controls do not consistently apply to all artifacts loaded for a model. A deployment that supplies --revision or --code-revision can still load dynamic code, GGUF files, image processors, retrieval side weights, or same-repository subfolder weights/config from an unpinned/default revision. This is a supply-chain integrity issue for pinned vLLM deployments. Operators can believe they… CWE-345Jun 22, 2026 | CVSS6.5v3.1 | EPSS0.21% | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-41523HIGH | vLLM: Security Check Bypass via assert Statement in Activation Function Loading Allows Arbitrary Code ExecutionvLLM is an inference and serving engine for large language models (LLMs). Prior to 0.22.0, an assert-based security check in vLLM's activation function loading allows any unauthenticated attacker to achieve arbitrary code execution on the server by publishing a malicious HuggingFace model, when vLLM runs in Python optimized mode (python -O or PYTHONOPTIMIZE=1). This vulnerability is fixed in 0.22.0. | CVSS7.5v3.1 | EPSS0.746% | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-54232HIGH | vLLM: Dependency Confusion Vulnerability in vLLM DockerfilevLLM is an inference and serving engine for large language models (LLMs). Prior to 0.22.1, the vLLM Dockerfile is vulnerable to a dependency confusion attack through the flashinfer-jit-cache package. The package is installed from a custom index (flashinfer.ai/whl/) using --extra-index-url, but the package name was not registered on PyPI, and UV_INDEX_STRATEGY="unsafe-best-match" is set globally. An attacker who registers flashinfer-jit-cache on PyPI with version 0.6.11.post2 can execute arbitrar… CWE-427Jun 22, 2026 | CVSS8.8v3.1 | EPSS0.563% | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-54233MEDIUM | vLLM: OOM Denial of Service via Audio Decompression BombvLLM is an inference and serving engine for large language models (LLMs). Prior to 0.23.1rc0, vLLM's /v1/audio/transcriptions endpoint limits compressed upload size but not decoded PCM output. A 25MB OPUS file expands to ~14.9GB of float32 PCM at decode time. This vulnerability is fixed in 0.23.1rc0. CWE-409Jun 22, 2026 | CVSS6.5v3.1 | EPSS0.422% | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-54236MEDIUM | vLLM: incomplete CVE-2026-22778 fix leaks PIL repr addresses via Anthropic routervLLM is an inference and serving engine for large language models (LLMs). Prior to 0.23.1rc0, the fix for CVE-2026-22778, which introduced a sanitize_message helper that strips object-repr memory addresses from error messages before they reach the client, is incomplete: several response paths echo str(exc) directly to clients without calling sanitize_message. The unsanitized sites include the Anthropic API router in vllm/entrypoints/anthropic/api_router.py (the POST /v1/messages and POST /v1/mes… | CVSS5.3v3.1 | EPSS0.823% | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware use1 Nuclei template | STIX |
CVE-2026-54235MEDIUM | vLLM: temperature=NaN and temperature=Infinity bypass validation and propagate to GPU kernelsvLLM is an inference and serving engine for large language models (LLMs). Prior to 0.23.1rc0, ll temperature validation gates use comparison operators (<, >), which silently evaluate to False for NaN and for positive Infinity in Python's IEEE 754 float semantics. Both values pass every guard and propagate to GPU sampling kernels, where they produce undefined behavior or CUDA errors that can crash the inference worker. This vulnerability is fixed in 0.23.1rc0. CWE-1287Jun 22, 2026 | CVSS6.9v4.0 | EPSS0.27% | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-48746CRITICAL | vLLM: OpenAI auth bypassvLLM 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 AuthenticationMiddleware. It allows to use the API without providing the configured VLLM_API_KEY or --api-key. This vulnerability is fixed in 0.22.0. | CVSS9.1v3.1 | EPSS1.15% | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-53923MEDIUM | vLLM GGUF Kernels: int64_t to int truncation of tensor dimensions causes GPU buffer overflowvLLM is an inference and serving engine for large language models (LLMs). From 0.5.5 until 0.23.1rc0, integer truncation of tensor dimensions in vLLM's GGUF dequantize kernels (csrc/quantization/gguf/gguf_kernel.cu) causes partial tensor processing. The output tensor is allocated at full size via torch::empty (uninitialized memory), but the dequantize CUDA kernel processes only a truncated number of elements. The unfilled portion of the output tensor retains whatever was previously in GPU memory… | CVSS5.3v4.0 | EPSS0.281% | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-12491MEDIUM | Vllm: vllm: image exif rotation & png trns transparency not normalized, causing mismatch between model input and expectationsA flaw was found in vLLM, an open-source library for large language model inference. This vulnerability arises from improper handling of image metadata, specifically EXIF orientation and PNG transparency (tRNS) data, during image processing. When images are converted to RGB, transparency information may be implicitly discarded or remapped, leading to unexpected rendering of transparent pixels and distortion of input content. This can result in the model misinterpreting image content, potentially… | CVSS4.8v3.1 | EPSS0.239% | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-9540MEDIUM | vllm-project vllm OpenAI-compatible Serving Path denial of serviceA vulnerability was identified in vllm-project vllm 0.19.0. This issue affects some unknown processing of the component OpenAI-compatible Serving Path. Such manipulation leads to denial of service. It is possible to launch the attack remotely. The exploit is publicly available and might be used. The pull request to fix this issue awaits acceptance. CWE-404May 26, 2026 | CVSS6.9v4.0 | EPSS0.427% | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-44223MEDIUM | vLLM: extract_hidden_states speculative decoding crashes server on any request with penalty parametersvLLM is an inference and serving engine for large language models (LLMs). From 0.18.0 to before 0.20.0, the extract_hidden_states speculative decoding proposer in vLLM returns a tensor with an incorrect shape after the first decode step, causing a RuntimeError that crashes the EngineCore process. The crash is triggered when any request in the batch uses sampling penalty parameters (repetition_penalty, frequency_penalty, or presence_penalty). A single request with a penalty parameter (e.g., "repe… | CVSS6.5v3.1 | EPSS0.367% | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-44222MEDIUM | vLLM: Remote DoS via Special-Token PlaceholdersvLLM is an inference and serving engine for large language models (LLMs). From 0.6.1 to before 0.20.0, there is a a Token Injection vulnerability in vLLM’s multimodal processing. Unauthenticated, text-only prompts that spell special tokens are interpreted as control. Image and video placeholder sequences supplied without matching data cause vLLM to index into empty grids during input-position computation, raising an unhandled IndexError and terminating the worker or degrading availability. Multi… CWE-129May 12, 2026 | CVSS6.5v3.1 | EPSS0.414% | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-34756MEDIUM | vLLM Affected by Unauthenticated OOM Denial of Service via Unbounded `n` Parameter in OpenAI API ServervLLM is an inference and serving engine for large language models (LLMs). From 0.1.0 to before 0.19.0, a Denial of Service vulnerability exists in the vLLM OpenAI-compatible API server. Due to the lack of an upper bound validation on the n parameter in the ChatCompletionRequest and CompletionRequest Pydantic models, an unauthenticated attacker can send a single HTTP request with an astronomically large n value. This completely blocks the Python asyncio event loop and causes immediate Out-Of-Memo… | CVSS6.5v3.1 | EPSS0.346% | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-34755MEDIUM | vLLM Affected by Denial of Service via Unbounded Frame Count in video/jpeg Base64 ProcessingvLLM is an inference and serving engine for large language models (LLMs). From 0.7.0 to before 0.19.0, the VideoMediaIO.load_base64() method at vllm/multimodal/media/video.py splits video/jpeg data URLs by comma to extract individual JPEG frames, but does not enforce a frame count limit. The num_frames parameter (default: 32), which is enforced by the load_bytes() code path, is completely bypassed in the video/jpeg base64 path. An attacker can send a single API request containing thousands of co… CWE-770Apr 6, 2026 | CVSS6.5v3.1 | EPSS0.378% | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-34753MEDIUM | vLLM affected by Server-Side Request Forgery (SSRF) in `download_bytes_from_url `vLLM is an inference and serving engine for large language models (LLMs). From 0.16.0 to before 0.19.0, a server-side request forgery (SSRF) vulnerability in download_bytes_from_url allows any actor who can control batch input JSON to make the vLLM batch runner issue arbitrary HTTP/HTTPS requests from the server, without any URL validation or domain restrictions. This can be used to target internal services (e.g. cloud metadata endpoints or internal HTTP APIs) reachable from the vLLM host. This … CWE-918Apr 6, 2026 | CVSS5.4v3.1 | EPSS0.246% | PoCs1 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |
CVE-2026-34760MEDIUM | vLLM: Downmix Implementation Differences as Attack Vectors Against Audio AI ModelsvLLM is an inference and serving engine for large language models (LLMs). From version 0.5.5 to before version 0.18.0, Librosa defaults to using numpy.mean for mono downmixing (to_mono), while the international standard ITU-R BS.775-4 specifies a weighted downmixing algorithm. This discrepancy results in inconsistency between audio heard by humans (e.g., through headphones/regular speakers) and audio processed by AI models (Which infra via Librosa, such as vllm, transformer). This issue has been… CWE-20Apr 2, 2026 | CVSS5.9v3.1 | EPSS0.267% | PoCs0 | SignalsNot listed in CISA KEVNo known ransomware useNo Nuclei templates | STIX |