vLLM versions 0.8.0 and later are vulnerable to an Out-of-Memory (OOM) Denial of Service (DoS) attack due to unbounded frame count processing in the `VideoMediaIO.load_base64()` method. When processing `video/jpeg` data URLs, the method splits the base64 data string on commas to extract individual JPEG frames without enforcing a frame count limit. An attacker can exploit this by crafting a single API request containing thousands of comma-separated base64-encoded JPEG frames in a data URL, causing the server to decode all frames into memory and crash due to excessive memory consumption. This vulnerability is reachable via the OpenAI-compatible chat completions API and does not require authentication.

Affected: PyPI: vllm

vLLM versions 0.8.0 and later are vulnerable to an Out-of-Memory (OOM) Denial of Service (DoS) attack due to unbounded frame count processing in the `VideoMediaIO.load_base64()` method. When processing `video/jpeg` data URLs, the method splits the base64 data string on commas to extract individual JPEG frames without enforcing a frame count limit. An attacker can exploit this by crafting a single API request containing thousands of comma-separated base64-encoded JPEG frames in a data URL, causing the server to decode all frames into memory and crash due to excessive memory consumption. This vulnerability is reachable via the OpenAI-compatible chat completions API and does not require authentication.

Affected: PyPI: vllm

vLLM versions 0.8.0 and later are vulnerable to an Out-of-Memory (OOM) Denial of Service (DoS) attack due to unbounded frame count processing in the `VideoMediaIO.load_base64()` method. When processing `video/jpeg` data URLs, the method splits the base64 data string on commas to extract individual JPEG frames without enforcing a frame count limit. An attacker can exploit this by crafting a single API request containing thousands of comma-separated base64-encoded JPEG frames in a data URL, causing the server to decode all frames into memory and crash due to excessive memory consumption. This vulnerability is reachable via the OpenAI-compatible chat completions API and does not require authentication.

python-transformers: python-transformers: Arbitrary code execution due to overridden trust_remote_code setting vllm: vLLM: Denial of Service via unbounded video frame processing aiohttp: AIOHTTP: Arbitrary code execution via untrusted input to CookieJar.load() vllm: vLLM: Arbitrary code execution via malicious HuggingFace model Pillow: python-pillow: Pillow: Arbitrary code execution via malicious PSD file processing vllm: vLLM: Denial of Service via malformed tensor shape in speculative decoding python-pyjwt: PyJWT: Authentication bypass due to forged JSON Web Tokens starlette: Starlette: Security restriction bypass via malformed HTTP Host header vllm: starlette: vLLM: Critical authentication bypass allows unauthorized API access Pillow: Pillow: Memory disclosure or denial of service via crafted McIdas AREA image python-pillow: Pillow: Denial of Service via excessive memory allocation when processing font files vllm: vLLM: Denial of Service via malformed speculative decoding workload starlette: Starlette: request.form() limits silently ignored for application/x-www-form-urlencoded enable DoS python-pillow: Pillow: Denial of Service via crafted BDF font file python-pillow: Pillow: Denial of Service via crafted GD 2.x image file vllm: vLLM: Denial of Service via adversarial regular expression in structured outputs API Pillow: Pillow: Native heap out-of-bounds write Pillow: Pillow: Denial of Service via out-of-bounds write in image processing Pillow: Pillow: Denial of service v…

Ein Angreifer kann mehrere Schwachstellen in vllm ausnutzen, um Sicherheitsmaßnahmen zu umgehen, einen Denial-of-Service-Zustand zu verursachen, Daten zu manipulieren oder vertrauliche Informationen offenzulegen.