CVE-2026-105754 vLLM: Scale-out disaggregated multimodal transport trusts caller-supplied features

vLLM is an inference and serving engine for large language models. Prior to 0.30.0, the /inference/v1/generate endpoint in the disaggregated scale-out path accepts caller-supplied tensors in the features.kwargsdata field, cache identifiers in the features.mmhashes field, ranges in the...

Improper Check for Unusual or Exceptional Conditions

Overview vllm is an A high-throughput and memory-efficient inference and serving engine for LLMs Affected versions of this package are vulnerable to Improper Check for Unusual or Exceptional Conditions via the disaggregated serving endpoint /inference/v1/generate in...

CVE-2026-100651

vLLM before 0.29.0 fails to enforce decoder prompt-length validation on the disaggregated serving endpoint /inference/v1/generate. When the request contains a 'features' multimodal payload, vllm/entrypoints/serve/disagg/serving.py builds a multimodal EngineInput directly from the caller-supplied...

CVE-2026-100651

vLLM before 0.29.0 fails to enforce decoder prompt-length validation on the disaggregated serving endpoint /inference/v1/generate. When the request contains a 'features' multimodal payload, vllm/entrypoints/serve/disagg/serving.py builds a multimodal EngineInput directly from the caller-supplied...

CVE-2026-100651

vLLM versions before 0.29.0 are vulnerable to a Denial of Service (DoS) due to a lack of decoder prompt-length validation on the disaggregated serving endpoint (/inference/v1/generate). When handling multimodal payloads (e.g., Nemotron Parse , Whisper , and FireRedLID ), the system fails to check...

CVE-2026-100651 vllm before 0.29.0 Denial of Service via Decoder Prompt Length Bypass

vLLM before 0.29.0 fails to enforce decoder prompt-length validation on the disaggregated serving endpoint /inference/v1/generate. When the request contains a 'features' multimodal payload, vllm/entrypoints/serve/disagg/serving.py builds a multimodal EngineInput directly from the caller-supplied...

EUVD-2026-87745

vLLM before 0.29.0 fails to enforce decoder prompt-length validation on the disaggregated serving endpoint /inference/v1/generate. When the request contains a 'features' multimodal payload, vllm/entrypoints/serve/disagg/serving.py builds a multimodal EngineInput directly from the caller-supplied...

CVE-2026-100651 vllm before 0.29.0 Denial of Service via Decoder Prompt Length Bypass

vLLM before 0.29.0 fails to enforce decoder prompt-length validation on the disaggregated serving endpoint /inference/v1/generate. When the request contains a 'features' multimodal payload, vllm/entrypoints/serve/disagg/serving.py builds a multimodal EngineInput directly from the caller-supplied...

CVE-2026-100651 vllm before 0.29.0 Denial of Service via Decoder Prompt Length Bypass

vLLM before 0.29.0 fails to enforce decoder prompt-length validation on the disaggregated serving endpoint /inference/v1/generate. When the request contains a 'features' multimodal payload, vllm/entrypoints/serve/disagg/serving.py builds a multimodal EngineInput directly from the caller-supplied...

CVE-2026-100651: Uncontrolled Resource Consumption

vLLM before 0.29.0 fails to enforce decoder prompt-length validation on the disaggregated serving endpoint /inference/v1/generate. When the request contains a 'features' multimodal payload, vllm/entrypoints/serve/disagg/serving.py builds a multimodal EngineInput directly from the caller-supplied...

PT-2026-99322

Name of the Vulnerable Software and Affected Versions vLLM versions prior to 0.29.0 Description The software fails to enforce decoder prompt-length validation on the disaggregated serving endpoint '/inference/v1/generate'. When a request includes a 'features' multimodal payload, the system builds...

SUSE CVE-2026-94627

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 exhaustion by submitting completion requests with multiple prompts,...

CVE-2026-93088 CVE-2026-93088

SGLang's multimodal generation runtime is vulnerable to unauthenticated arbitrary code execution because the disaggregated-diffusion orchestrator's DiffusionServer binds an unauthenticated ZeroMQ ROUTER socket to a network interface and passes the final frame of received multipart messages direct...

CVE-2026-93088 CVE-2026-93088

SGLang's multimodal generation runtime is vulnerable to unauthenticated arbitrary code execution because the disaggregated-diffusion orchestrator's DiffusionServer binds an unauthenticated ZeroMQ ROUTER socket to a network interface and passes the final frame of received multipart messages direct...

EUVD-2026-84221

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 exhaustion by submitting completion requests with multiple prompts,...

Reachable Assertion

Overview vllm is an A high-throughput and memory-efficient inference and serving engine for LLMs Affected versions of this package are vulnerable to Reachable Assertion via an assertion failure in NixlBaseConnectorWorker.applyprefixcaching, which fails to properly validate block counts across...

PYSEC-2026-4005

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 deployments. Attackers can trigger an assertion failure in...

CVE-2026-94626

vLLM through 0.29.0 fails to validate the tpsize parameter in kvtransferparams on OpenAI-compatible completion endpoints, allowing attackers to allocate unbounded memory. Attackers can supply arbitrary tpsize values in prefill/decode disaggregated deployments to exhaust memory and trigger kernel...

CVE-2026-94627

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 exhaustion by submitting completion requests with multiple prompts,...

PYSEC-2026-4004

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 kvtransferparams dictionary entries to trigger an uncaught KeyError in EngineCore scheduling,...