What is the CVSS score of CVE-2026-47615?

CVE-2026-47615 has a CVSS 3.1 base score of 7.5 (High). CVSS vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N. EPSS exploitation probability: 0.3%.

Which versions of NVIDIA Dynamo are affected by CVE-2026-47615?

NVIDIA Dynamo for Linux (all versions through v1.1.0) contains an unauthenticated server-side request forgery vulnerability that allows remote attackers to access internal network resources and…

How to fix or mitigate CVE-2026-47615?

Within 24 hours, inventory all NVIDIA Dynamo deployments and identify which instances are network-exposed or process untrusted inference payloads; isolate critical instances or restrict network access immediately. Within 7 days, implement network segmentation to separate Dynamo servers from sensitive internal resources, deploy outbound traffic monitoring to detect exploitation attempts via anomalous HTTP requests, and configure firewall rules blocking unnecessary outbound connections from Dynamo hosts. Within 30 days, monitor NVIDIA security advisories for patch availability, test deployment procedures in non-production, and establish a rapid-deployment plan for fixes as released.

NVIDIA Dynamo CVE-2026-47615

EUVD-2026-52809 HIGH

Server-Side Request Forgery (SSRF) (CWE-918)

2026-08-04 nvidia

GHSA-hc3w-j3wp-q93q

Information Disclosure SSRF Nvidia Dynamo

7.5

CVSS 3.1 · Vendor: nvidia

Severity by source

Vendor (nvidia) PRIMARY

7.5 HIGH

AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N

vuln.today AI

7.5 HIGH

Network-reachable inference API with no credential requirement; SSRF yields confidentiality impact only, no integrity or availability consequence.

3.1 AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N

4.0 AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:N/VA:N/SC:N/SI:N/SA:N

Primary rating from Vendor (nvidia).

CVSS VectorVendor: nvidia

Attack Vector

Network

Attack Complexity

Low

Privileges Required

None

User Interaction

None

Scope

Unchanged

Confidentiality

High

Integrity

None

Availability

None

Lifecycle Timeline

Analysis Generated

Aug 04, 2026 - 18:05 vuln.today

CVE Published

Aug 04, 2026 - 17:30 cve.org

HIGH 7.5

DescriptionCVE.org

NVIDIA Dynamo for Linux contains a vulnerability where an attacker may cause server-side request forgery by supplying a crafted URL in a multimodal request. A successful exploit of this vulnerability might lead to information disclosure.

AnalysisAI

Server-side request forgery in NVIDIA Dynamo for Linux (all versions through v1.1.0) allows unauthenticated remote attackers to force the inference server to issue arbitrary HTTP requests by embedding a crafted URL inside a multimodal inference request payload. Successful exploitation exposes internal network resources reachable from the Dynamo host, producing high-severity information disclosure without requiring any credentials or user interaction. No public exploit code or CISA KEV listing has been identified at time of analysis, though the PR:N/AC:L attack surface on AI serving endpoints that may be network-exposed warrants prompt remediation.

Technical ContextAI

NVIDIA Dynamo (CPE: cpe:2.3:a:nvidia:dynamo:*:*:*:*:*:*:*:*) is a distributed AI inference serving framework designed for deploying large language models at scale on Linux. Multimodal inference requests allow clients to supply URLs pointing to external media assets (images, audio, video) that the server fetches server-side for model input processing. CWE-918 (Server-Side Request Forgery) arises when this URL-fetching logic fails to validate or restrict destination hosts, enabling the server to act as an unauthenticated HTTP proxy. An attacker can supply URLs targeting RFC-1918 addresses, cloud metadata endpoints (e.g., 169.254.169.254), or other internal services ordinarily inaccessible from the attacker's vantage point. The CVSS vector AV:N/AC:L/PR:N/UI:N confirms the issue is reachable over the network without credentials or special triggering conditions, consistent with a publicly accessible inference API endpoint.

RemediationAI

Upgrade NVIDIA Dynamo to a version beyond v1.1.0 as directed by the NVIDIA Product Security advisory at https://github.com/NVIDIA/product-security/tree/main/2026/5842; the exact patched release version is not independently confirmed from available data, so consult the advisory for the specific target version. If an immediate upgrade is not possible, the following compensating controls reduce exposure: restrict egress from the Dynamo server at the host or network firewall level to block requests to RFC-1918 address space, cloud metadata endpoints (169.254.169.254, fd00:ec2::254), and any internal services that should not be reachable from the inference server - this directly limits the information an attacker can extract via SSRF, at the trade-off of potentially breaking legitimate multimodal URL-fetch functionality if internal asset stores are used. Additionally, place Dynamo inference endpoints behind an API gateway or reverse proxy that enforces authentication before forwarding requests, reducing the PR:N attack surface even if the underlying SSRF is unpatched. Network segmentation to isolate Dynamo hosts from sensitive internal services is also advisable regardless of patching status.