What is the CVSS score of EUVD-2026-52810?

EUVD-2026-52810 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 EUVD-2026-52810?

NVIDIA Dynamo versions through v1.1.0 contain an unauthenticated server-side request forgery vulnerability (CVSS 7.5) that allows attackers to redirect server traffic to arbitrary internal and cloud…

How to fix or mitigate EUVD-2026-52810?

Within 24 hours, inventory all NVIDIA Dynamo deployments running v1.1.0 or earlier on Linux and restrict network access to inference endpoints to trusted clients only. Within 7 days, implement network segmentation isolating Dynamo servers from cloud metadata services and internal resources; disable AWS/GCP/Azure metadata endpoints at the instance level where possible. Within 30 days, monitor NVIDIA advisories for patch release and plan immediate upgrade; until patched versions are available, enforce strict URL allowlisting on all media input parameters and consider temporarily disabling the multimodal media fetcher if not critical to production operations.

NVIDIA Dynamo EUVD-2026-52810

| CVE-2026-47616 HIGH

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

2026-08-04 nvidia

GHSA-65pc-qfwh-rrgx

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-accessible unauthenticated media fetcher with no user interaction required; SSRF impact is confidentiality-only with no integrity or availability consequence confirmed.

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 in the multimodal media fetcher where an attacker may cause server-side request forgery. A successful exploit of this vulnerability might lead to information disclosure.

AnalysisAI

Server-side request forgery in NVIDIA Dynamo's multimodal media fetcher (Linux, versions through v1.1.0) allows unauthenticated remote attackers to issue arbitrary server-side HTTP requests by supplying malicious URLs as media inputs during inference. The CVSS 3.1 vector confirms network-accessible, zero-authentication exploitation with high confidentiality impact, scoring 7.5. No active exploitation or public exploit code has been identified at time of analysis, but SSRF in cloud-hosted AI inference infrastructure carries meaningful risk due to cloud metadata service exposure.

Technical ContextAI

NVIDIA Dynamo is an AI inference serving framework for Linux, designed to handle multimodal inference workloads including image, audio, and other media inputs. The vulnerable component is the multimodal media fetcher - a module that retrieves user-supplied media URLs server-side to support inference tasks. CWE-918 (Server-Side Request Forgery) describes vulnerabilities where insufficient validation of user-controlled URLs allows an attacker to manipulate the server into issuing HTTP requests to unintended destinations. In cloud-deployed environments, this commonly enables access to instance metadata services (e.g., AWS IMDSv1 at 169.254.169.254), internal microservices, or other network resources not reachable by the attacker directly. The affected CPE is cpe:2.3:a:nvidia:dynamo:*:*:*:*:*:*:*:*, covering all versions up to and including v1.1.0 on Linux.

RemediationAI

Upgrade NVIDIA Dynamo beyond v1.1.0 per the NVIDIA product security advisory at https://github.com/NVIDIA/product-security/tree/main/2026/5842; note that the exact fixed release version is not stated in available data and should be confirmed directly from the advisory before deployment. If an immediate upgrade is not feasible, restrict network access to Dynamo inference endpoints at the perimeter so only trusted, authenticated clients can submit requests - this removes the unauthenticated attack surface documented in the CVSS vector. As an additional compensating control, enforce cloud metadata service protection: on AWS, mandate IMDSv2 and disable IMDSv1 to prevent credential theft via metadata endpoint SSRF; on all platforms, apply host-level firewall rules blocking outbound HTTP from the Dynamo process to 169.254.0.0/16 and internal RFC-1918 ranges not required for legitimate operation. These controls reduce SSRF impact but do not eliminate the underlying vulnerability - patching remains the required remediation.