130 matches found
VLLM: Out-of-vocabulary Prompt Token Id Passes Input Validation And Triggers A CUDA Device-side Assert (engine Crash / D
For a model whose vocabulary size is an exact multiple of the embedding padding size 64 — e.g. any Llama-2-family model, vocab 32000 — a single request carrying a pre-tokenized prompt with token id equal to the vocab size 32000 passes vLLM's input-id validation but is one past the last embedding...
CVE-2026-115916 vLLM: Out-of-vocabulary prompt token id passes input validation and triggers a CUDA device-side assert (engine crash / DoS)
vLLM is an inference and serving engine for large language models. Prior to 0.32.0, vllm/tokenizers/hf.py computes maxtokenid with an off-by-one upper bound, allowing a pre-tokenized prompt ID equal to the vocabulary size to pass validation. For a tensor-parallel size of one and an exact...
vLLM input validation error vulnerability
Versions vLLM 0.20.0 through 0.29.0 contain an input validation error vulnerability. This vulnerability arises from sending inconsistent multimodal metadata to the /split/inference/v1/generate endpoint. By replaying valid multimodal entries that contain repeated or overlapping mmplaceholders with...
CVE-2026-114050 vLLM before 0.31.0 Remote Engine Crash via Forged Multimodal UUID
vLLM before 0.31.0 uses a client-supplied multimodal uuid verbatim as cache identity with no consistency check on cache hits, allowing engine crashes. Unauthenticated attackers can desynchronize the P0/P1 processor caches to trigger a fatal token-placeholder mismatch, killing the entire serving...
CVE-2026-114050 vLLM before 0.31.0 Remote Engine Crash via Forged Multimodal UUID
vLLM before 0.31.0 uses a client-supplied multimodal uuid verbatim as cache identity with no consistency check on cache hits, allowing engine crashes. Unauthenticated attackers can desynchronize the P0/P1 processor caches to trigger a fatal token-placeholder mismatch, killing the entire serving...
CVE-2026-114050
Versions of vLLM before 0.31.0 are vulnerable to a remote engine crash due to insufficient verification of data authenticity (CWE-345 ). The software uses a client-supplied multimodal uuid as a cache identity without performing any consistency checks on cache hits. This allows unauthenticated att...
CVE-2026-114050 vLLM before 0.31.0 Remote Engine Crash via Forged Multimodal UUID
vLLM before 0.31.0 uses a client-supplied multimodal uuid verbatim as cache identity with no consistency check on cache hits, allowing engine crashes. Unauthenticated attackers can desynchronize the P0/P1 processor caches to trigger a fatal token-placeholder mismatch, killing the entire serving...
SUSE CVE-2026-93592
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 negative token ID triggers a CUDA device-side assertion that...
PYSEC-0000-CVE-2026-93592
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 negative token ID triggers a CUDA device-side assertion that...
CVE-2026-92971 InternLM LMDeploy through 0.17.0 Assertion Denial of Service
InternLM LMDeploy through 0.17.0 contains a reachable assertion vulnerability in the DistServe decode migration loop that allows unauthenticated attackers to terminate the inference engine. Attackers can submit a migrationrequest with an empty remoteblockids list to trigger an AssertionError that...
Astra Linux – Vulnerability in Firefox
It was possible to mutate a JavaScript object in such a way that the JIT compiler could crash while tracing it. This vulnerability affects Firefox versions less than 125...
Linux Distros Unpatched Vulnerability : CVE-2026-57222
The Linux/Unix host has one or more packages installed that are impacted by a vulnerability without a vendor supplied patch available. - Suricata is a network Intrusion Detection System, Intrusion Prevention System and Network Security Monitoring engine. Prior to 7.0.17 and 8.0.6, crafted IPv4 an...
OESA-2026-3584 wireshark security update
Wireshark allows you to examine protocol data stored in files or as it is captured from wired or wireless WiFi or Bluetooth networks, USB devices, and many other sources. It supports dozens of protocol capture file formats and understands more than a thousand protocols. Security Fixes: C12.22...
PT-2026-60608
Wazuh is a free and open source platform used for threat prevention, detection, and response. In versions 1.0.0 and above, prior to 4.14.5, a heap buffer overflow in wazuh-analysisd allows an unauthenticated remote attacker to crash the Wazuh manager's analysis engine, causing complete loss of SI...
vLLM vulnerable to DoS with incorrect shape of multimodal embedding inputs
SummaryUsers can crash the vLLM engine serving multimodal models by passing multimodal embedding inputs with correct ndim but incorrect shape e.g. hidden dimension is wrong, regardless of whether the model is intended to support such inputs as defined in the Supported Models page.The issue has...
PYSEC-2026-145
vLLM is an inference and serving engine for large language models LLMs. From to before 0.20.0, the extracthiddenstates 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 ...
CVE-2026-44223
vLLM is an inference and serving engine for large language models LLMs. From 0.18.0 to before 0.20.0, the extracthiddenstates 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...
CVE-2026-31443 dmaengine: idxd: Fix crash when the event log is disabled
In the Linux kernel, the following vulnerability has been resolved: dmaengine: idxd: Fix crash when the event log is disabled If reporting errors to the event log is not supported by the hardware, and an error that causes Function Level Reset FLR is received, the driver will try to restore the...
CVE-2026-33285 LiquidJS: memoryLimit Bypass through Negative Range Values Leads to Process Crash
LiquidJS is a Shopify / GitHub Pages compatible template engine in pure JavaScript. Prior to version 10.25.1, LiquidJS's memoryLimit security mechanism can be completely bypassed by using reverse range expressions e.g., 100000000..1, allowing an attacker to allocate unlimited memory. Combined wit...
CVE-2026-33285 LiquidJS: memoryLimit Bypass through Negative Range Values Leads to Process Crash
LiquidJS is a Shopify / GitHub Pages compatible template engine in pure JavaScript. Prior to version 10.25.1, LiquidJS's memoryLimit security mechanism can be completely bypassed by using reverse range expressions e.g., 100000000..1, allowing an attacker to allocate unlimited memory. Combined wit...