CVE-2026-73556 vLLM: ReDoS via structured_outputs.regex in the lm-format-enforcer backend (no compile timeout) — missed sibling of CVE-2026-55574

vLLM is an inference and serving engine for large language models. Prior to 0.26.0, the structuredoutputs.regex parameter in vllm/v1/structuredoutput/backendlmformatenforcer.py is passed to lmformatenforcer.RegexParser without compileregexwithtimeout or validation in...

CVE-2026-73556 vLLM: ReDoS via structured_outputs.regex in the lm-format-enforcer backend (no compile timeout) — missed sibling of CVE-2026-55574

vLLM is an inference and serving engine for large language models. Prior to 0.26.0, the structuredoutputs.regex parameter in vllm/v1/structuredoutput/backendlmformatenforcer.py is passed to lmformatenforcer.RegexParser without compileregexwithtimeout or validation in...

CVE-2026-73555 vLLM: Unauthenticated Internal Path and Username Disclosure via Validation Error Messages

vLLM is an inference and serving engine for large language models. Prior to 0.26.0, the validationexceptionhandler in vllm/entrypoints/openai/serverutils.py converts FastAPI RequestValidationError objects with strexc, and sanitizemessage in vllm/entrypoints/utils.py does not remove traceback-styl...

CVE-2026-73555

vLLM is an inference and serving engine for large language models. Prior to 0.26.0, the validationexceptionhandler in vllm/entrypoints/openai/serverutils.py converts FastAPI RequestValidationError objects with strexc, and sanitizemessage in vllm/entrypoints/utils.py does not remove traceback-styl...

CVE-2026-73555

CVE-2026-73555 affects vLLM prior to 0.26.0. The issue arises in the validation_exception_handler (vllm/entrypoints/openai/server_utils.py) which converts FastAPI RequestValidationError objects with str(exc), and in sanitize_message (vllm/entrypoints/utils.py) which does not remove traceback-styl...

CVE-2026-73555 vLLM: Unauthenticated Internal Path and Username Disclosure via Validation Error Messages

vLLM is an inference and serving engine for large language models. Prior to 0.26.0, the validationexceptionhandler in vllm/entrypoints/openai/serverutils.py converts FastAPI RequestValidationError objects with strexc, and sanitizemessage in vllm/entrypoints/utils.py does not remove traceback-styl...

PT-2026-71673

Name of the Vulnerable Software and Affected Versions vLLM versions prior to 0.27.0 Description An integer overflow occurs in the calculation blockIdx.x 2 d within the activation kernels.cu file. This flaw can cause the act and mul kernel function to consume input from another user within the sam...

PT-2026-71670

Name of the Vulnerable Software and Affected Versions vLLM versions prior to 0.26.0 Description An issue exists where the validation exception handler in vllm/entrypoints/openai/server utils.py converts FastAPI RequestValidationError objects using strexc, and the sanitize message function in...

PT-2026-71675

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 liststr or listlistint, prompt to seq in vllm/renderers/inputs/preprocess.py and...

PT-2026-71672

Name of the Vulnerable Software and Affected Versions vLLM versions 0.20.2rc0 through 0.25.x Description A race condition exists in the safe load prompt embeds function within vllm/renderers/embed utils.py when enable prompt embeds is enabled. The issue occurs because torch.sparse.check sparse...

PT-2026-71671

Name of the Vulnerable Software and Affected Versions vLLM versions prior to 0.26.0 Description An unauthenticated request to the '/v1/completions' endpoint can cause a denial of service by consuming a CPU core and stalling the structured-output engine path. This occurs because the structured...

CVE-2026-27765

Improper input validation for some vLLM Hardware Plugin for IntelR GaudiR software before version 0.16.0 within Ring 3: User Applications may allow a denial of service. Authorized adversary with an authenticated user combined with a low complexity attack may enable denial of service. This result...

EUVD-2026-56522

Improper input validation for some vLLM Hardware Plugin for IntelR GaudiR software before version 0.16.0 within Ring 3: User Applications may allow a denial of service. Authorized adversary with an authenticated user combined with a low complexity attack may enable denial of service. This result...

CVE-2026-27765

Improper input validation for some vLLM Hardware Plugin for IntelR GaudiR software before version 0.16.0 within Ring 3: User Applications may allow a denial of service. Authorized adversary with an authenticated user combined with a low complexity attack may enable denial of service. This result...

CVE-2026-27765

The CVE affects some vLLM Hardware Plugin for Intel) Gaudi software prior to 0.16.0, where improper input validation in Ring 3 User Applications may allow a local, authenticated user with low complexity to cause a denial of service. The attack requires no user interaction and leverages local acce...

CVE-2026-27765

Improper input validation for some vLLM Hardware Plugin for IntelR GaudiR software before version 0.16.0 within Ring 3: User Applications may allow a denial of service. Authorized adversary with an authenticated user combined with a low complexity attack may enable denial of service. This result...

vLLM Hardware Plugin for Intel® Gaudi® Software Advisory

Summary: A potential security vulnerability for some vLLM Hardware Plugin for Intel® Gaudi® Software may allow denial of service. Intel is releasing software updates to mitigate this potential vulnerability. Vulnerability Details: CVEID: CVE-2026-27765 Description: Improper input validation for...

PT-2026-70332

Improper input validation for some vLLM Hardware Plugin for IntelR GaudiR software before version 0.16.0 within Ring 3: User Applications may allow a denial of service. Authorized adversary with an authenticated user combined with a low complexity attack may enable denial of service. This result...

PYSEC-2026-3542 vLLM has Remote DoS via Invalid Recovered Token Reinjection

Summary A frontend-legal multi-request speculative workload can make vLLM produce an out-of-vocabulary recovered token equal to vocabsize, convert that value to -1 when choosing the next live token for a request, and then feed that -1 back into the next drafter input ids. On Qwen3 GPTQ this reach...

GHSA-33CG-GXV8-3P8G vLLM denial of service via prompt embeds on M-RoPE models

Summary Short summary of the problem. Make the impact and severity as clear as possible. For example: An unsafe deserialization vulnerability allows any unauthenticated user to execute arbitrary code on the server. Sending a pure prompt embeds payload in a /v1/completions request with a model usi...