Security Vulnerabilities
vLLM is an inference and serving engine for large language models. From 0.16.0 until 0.32.0, the /v1/responses endpoint allows an authenticated caller to place an attacker-controlled template in chat_template_kwargs[chat_template] when no server template is configured. ChatParams.get_apply_chat_template_kwargs() preserves that value without enforcing trust_request_chat_template, so the request can bypass the intended trust policy and cause CPU or memory exhaustion through Jinja rendering. This issue is fixed in version 0.32.0.
CVSS Score
6.5
EPSS Score
0.004
Published
2026-10-26
vLLM is an inference and serving engine for large language models. From 0.11.1 until 0.32.0, the DeepSeek-OCR, DeepSeek-OCR-2, and Unlimited-OCR image path allows a remote client that can reach the OpenAI-compatible API to submit a legal image whose normalized pixels sum to zero when that image is alone in its per-step encoder batch. _parse_and_validate_image_input returns no embedding, embed_multimodal propagates that value, and sanity_check_mm_encoder_outputs triggers a fatal assertion that shuts down the engine for all users. This issue is fixed in version 0.32.0.
CVSS Score
6.5
EPSS Score
0.004
Published
2026-10-26
vLLM is an inference and serving engine for large language models. Prior to 0.32.0, ImageMediaIO passes attacker-controlled image bytes to Pillow through Image.open(BytesIO(data)) without restricting decoder selection, and the MIME label does not constrain the selected format. Image-capable request paths through parse_image() and fetch_image() can therefore dispatch an EPS payload to host components. A host with Ghostscript affected by CVE-2024-29510 can execute commands with the vLLM process's privileges, while environments using Pillow 12.0.0 through 12.2.0 can trigger CVE-2026-59203 and exhaust media-loading workers. Deployments without the applicable vulnerable dependency are not affected by that corresponding impact. This issue is fixed in version 0.32.0.
CVSS Score
8.1
EPSS Score
0.007
Published
2026-10-26
A security flaw has been discovered in 2K Games Sid Meier's Civilization VI 1.0.12.68. The affected element is the function loadstring of the component Lua sandbox. Performing a manipulation results in type confusion. The attack may be initiated remotely. The exploit has been released to the public and may be used for attacks. The patch is named Hotfix November 2026. The vendor was contacted early about this disclosure.
CVSS Score
2.1
EPSS Score
0.004
Published
2026-10-26
vLLM is an inference and serving engine for large language models. Prior to 0.32.0, the Phi-4 multimodal audio encoder in phi4mm_audio.py and phi4mm_utils.py uses eager attention by default, allowing a remote client to submit one legal long audio clip that causes MultiHeadedAttention.forward_attention to materialize and retain full attention matrices in self.attn. The scaled dot-product attention branch does not materialize these matrices. The resulting quadratic GPU-memory use can cause an out-of-memory condition and terminate EngineCore for all users. This issue is fixed in version 0.32.0.
CVSS Score
6.5
EPSS Score
0.003
Published
2026-10-26
vLLM is an inference and serving engine for large language models. From 0.19.0 until 0.32.0, authenticated callers can supply a request chat template when no server-side --chat-template is configured, or a malicious model repository can provide a template that is compiled and rendered by tokenizer.apply_chat_template() without a wall-clock timeout, output-size limit, or iteration bound. Rendering runs in a ThreadPoolExecutor with one worker by default, so a nonterminating render can block /v1/responses and other chat-template users after the request ends, while an output-expanding render can exhaust server memory. Harmony models are not affected because they do not use this Jinja path. This issue is fixed in version 0.32.0.
CVSS Score
6.5
EPSS Score
0.003
Published
2026-10-26
vLLM is an inference and serving engine for large language models. Prior to 0.32.0, an authenticated caller can place forged shared-memory address and monotonic_id values in public multimodal features.kwargs_data. In scale-out deployments with shared-memory multimodal caching enabled, workers trust those values and pass attacker-selected bytes to pickle.loads(), allowing arbitrary Python execution even when insecure serialization is disabled. This issue is fixed in version 0.32.0.
CVSS Score
7.5
EPSS Score
0.005
Published
2026-10-26
Mattermost versions 11.7.x <= 11.7.10 fail to validate reserved system post types on the scheduled post creation and publication paths which allows an authenticated channel member to publish messages rendering as trusted system events via the scheduled post endpoint.. Mattermost Advisory ID: MMSA-2026-00732
CVSS Score
3.5
EPSS Score
0.002
Published
2026-10-26
Mattermost versions 11.7.x <= 11.7.10 fail to apply outbound internal-connection filtering to the OpenID discovery endpoint request which allows a System Administrator to determine whether internal addresses and ports are reachable from the server via an unauthenticated login request that triggers the discovery fetch.. Mattermost Advisory ID: MMSA-2026-00774
CVSS Score
2.7
EPSS Score
0.003
Published
2026-10-26
Mattermost versions 11.9.x <= 11.9.1, 11.8.x <= 11.8.5, 11.7.x <= 11.7.10, 11.10.x <= 11.10.1 Mattermost fail to apply allocation limits to the byte-string decode paths in its WebSocket msgpack decoder, which allows an authenticated remote attacker to cause excessive server memory allocation via a crafted binary WebSocket frame declaring a Bin32 payload length of up to 4,294,967,295 bytes.. Mattermost Advisory ID: MMSA-2026-00768
CVSS Score
6.5
EPSS Score
0.005
Published
2026-10-26