https://db.gcve.eu Contains only the most 10 recent entries. http://www.rssboard.org/rss-specification python-feedgen en Fri, 09 Oct 2026 13:01:42 +0000 https://db.gcve.eu/vuln/fkie_cve-2026-105760 <p>vLLM is an inference and serving engine for large language models. Prior to 0.30.0, a caller can use the request-level media_io_kwargs field to select the GLMGA video backend and supply large values for the fps and max_frames options without a strict work ceiling. GLMGA constructs and deduplicates an attacker-sized pre-decode frame-index list, allowing a compact request and tiny valid video to consume disproportionate CPU time and memory in the shared media-loading executor. This issue is fixed in version 0.30.0.</p> <p>vLLM is an inference and serving engine for large language models. Prior to 0.30.0, a caller can use the request-level media_io_kwargs field to select the GLMGA video backend and supply large values for the fps and max_frames options without a strict work ceiling. GLMGA constructs and deduplicates an attacker-sized pre-decode frame-index list, allowing a compact request and tiny valid video to consume disproportionate CPU time and memory in the shared media-loading executor. This issue is fixed in version 0.30.0.</p> https://db.gcve.eu/vuln/fkie_cve-2026-105760 https://db.gcve.eu/vuln/ghsa-58v5-2m8f-94pr <p><strong>Affected:</strong> PyPI: vllm</p> <p>## Summary</p> <p>The OpenAI-compatible chat endpoint accepts request-level video loader options through `media_io_kwargs`. A caller can select the GLMGA sampler and provide large `fps` and `max_frames` values. GLMGA first constructs an attacker-sized Python index list and then deduplicates it, even when the supplied video contains only a few frames. This work occurs in the shared media-loading executor before frame reads, allowing a tiny valid video and compact JSON options to consume disproportionate CPU time and memory and delay unrelated media requests.</p> <p>The demonstrated impact is partial denial of service. No memory corruption, data disclosure, or code execution is claimed.</p> <p>## Affected configuration</p> <p>The server must expose chat completions for a video-capable model and accept request-level `media_io_kwargs`. The model does not need to use GLMGA by default because the request value overrides the model&#39;s loader mapping. The proof of concept uses an inline two-frame video and explicitly selects the OpenCV decoder, so no external media host or GPU decoder is required.</p> <p>The request is reachable without vLLM credentials when no API key is configured. With API-key authentication enabled, any caller holding a valid key can reach the same path.</p> <p>## Attack surface</p> <p>A remote caller submits a valid chat-completion request with a small video and the following request-level options:</p> <p>```json { &#34;media_io_kwargs&#34;: { &#34;video&#34;: { &#34;video_backend&#34;: &#34;glmga&#34;, &#34;backend&#34;: &#34;opencv&#34;,…</p> <p><strong>Affected:</strong> PyPI: vllm</p> <p>## Summary</p> <p>The OpenAI-compatible chat endpoint accepts request-level video loader options through `media_io_kwargs`. A caller can select the GLMGA sampler and provide large `fps` and `max_frames` values. GLMGA first constructs an attacker-sized Python index list and then deduplicates it, even when the supplied video contains only a few frames. This work occurs in the shared media-loading executor before frame reads, allowing a tiny valid video and compact JSON options to consume disproportionate CPU time and memory and delay unrelated media requests.</p> <p>The demonstrated impact is partial denial of service. No memory corruption, data disclosure, or code execution is claimed.</p> <p>## Affected configuration</p> <p>The server must expose chat completions for a video-capable model and accept request-level `media_io_kwargs`. The model does not need to use GLMGA by default because the request value overrides the model&#39;s loader mapping. The proof of concept uses an inline two-frame video and explicitly selects the OpenCV decoder, so no external media host or GPU decoder is required.</p> <p>The request is reachable without vLLM credentials when no API key is configured. With API-key authentication enabled, any caller holding a valid key can reach the same path.</p> <p>## Attack surface</p> <p>A remote caller submits a valid chat-completion request with a small video and the following request-level options:</p> <p>```json { &#34;media_io_kwargs&#34;: { &#34;video&#34;: { &#34;video_backend&#34;: &#34;glmga&#34;, &#34;backend&#34;: &#34;opencv&#34;,…</p> https://db.gcve.eu/vuln/ghsa-58v5-2m8f-94pr