CVE-2026-0599

{"id": "CVE-2026-0599", "cveTags": [], "metrics": {"cvssMetricV30": [{"type": "Secondary", "source": "security@huntr.dev", "cvssData": {"scope": "UNCHANGED", "version": "3.0", "baseScore": 7.5, "attackVector": "NETWORK", "baseSeverity": "HIGH", "vectorString": "CVSS:3.0/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H", "integrityImpact": "NONE", "userInteraction": "NONE", "attackComplexity": "LOW", "availabilityImpact": "HIGH", "privilegesRequired": "NONE", "confidentialityImpact": "NONE"}, "impactScore": 3.6, "exploitabilityScore": 3.9}]}, "published": "2026-02-02T11:16:17.773", "references": [{"url": "https://github.com/huggingface/text-generation-inference/commit/24ee40d143d8d046039f12f76940a85886cbe152", "source": "security@huntr.dev"}, {"url": "https://huntr.com/bounties/1d3f2085-666c-4441-b265-22f6f7d8d9cd", "source": "security@huntr.dev"}], "vulnStatus": "Deferred", "weaknesses": [{"type": "Primary", "source": "security@huntr.dev", "description": [{"lang": "en", "value": "CWE-400"}]}], "descriptions": [{"lang": "en", "value": "A vulnerability in huggingface/text-generation-inference version 3.3.6 allows unauthenticated remote attackers to exploit unbounded external image fetching during input validation in VLM mode. The issue arises when the router scans inputs for Markdown image links and performs a blocking HTTP GET request, reading the entire response body into memory and cloning it before decoding. This behavior can lead to resource exhaustion, including network bandwidth saturation, memory inflation, and CPU overutilization. The vulnerability is triggered even if the request is later rejected for exceeding token limits. The default deployment configuration, which lacks memory usage limits and authentication, exacerbates the impact, potentially crashing the host machine. The issue is resolved in version 3.3.7."}, {"lang": "es", "value": "Una vulnerabilidad en huggingface/text-generation-inference versi\u00f3n 3.3.6 permite a atacantes remotos no autenticados explotar la obtenci\u00f3n ilimitada de im\u00e1genes externas durante la validaci\u00f3n de entrada en modo VLM. El problema surge cuando el router escanea las entradas en busca de enlaces de imagen Markdown y realiza una solicitud HTTP GET bloqueante, leyendo todo el cuerpo de la respuesta en la memoria y clon\u00e1ndolo antes de decodificar. Este comportamiento puede llevar al agotamiento de recursos, incluyendo la saturaci\u00f3n del ancho de banda de la red, la inflaci\u00f3n de memoria y la sobreutilizaci\u00f3n de la CPU. La vulnerabilidad se activa incluso si la solicitud es posteriormente rechazada por exceder los l\u00edmites de tokens. La configuraci\u00f3n de despliegue predeterminada, que carece de l\u00edmites de uso de memoria y autenticaci\u00f3n, exacerba el impacto, pudiendo provocar el bloqueo de la m\u00e1quina anfitriona. El problema est\u00e1 resuelto en la versi\u00f3n 3.3.7."}], "lastModified": "2026-04-15T14:34:27.800", "sourceIdentifier": "security@huntr.dev"}