CVE-2026-63086 | text-generation-inference through 3.3.7 contains a server-si… | CVSS 8.6 HIGH

Description

text-generation-inference through 3.3.7 contains a server-side request forgery (SSRF) vulnerability in the OpenAI-compatible multimodal chat completions endpoint that allows unauthenticated network attackers to coerce the server into issuing arbitrary HTTP GET requests by supplying a crafted image_url value in chat message content. The fetch_image function in router/src/validation.rs performs no validation of private, loopback, link-local, or cloud metadata target addresses, and the reqwest HTTP client follows redirects by default, enabling attackers to bypass scheme checks via redirect chains to reach internal services and cloud instance-metadata endpoints for internal port scanning and credential theft.

Metadata

CVE ID: CVE-2026-63086
State: PUBLISHED
Assigner: VulnCheck
Reserved: 2026-07-15 15:45 UTC
Published: 2026-07-16 15:59 UTC
Last updated: 2026-07-16 17:41 UTC
Primary CWE: CWE-918
Server-Side Request Forgery (SSRF)
Vendor / Product: huggingface / text-generation-inference
Sources: cve.org · NVD

Severity & Metrics

8.6 HIGH CVSS 3.1

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

SSVC — CISA Coordinator

Exploitation: PoC
Automatable: yes
Tech. Impact: partial

Affected products (1)

Vendor	Product	Platform	Versions
huggingface	text-generation-inference	—	0 ≤ 3.3.7

Weakness (CWE)

CWE	Source	Description
CWE-918	cna	Server-Side Request Forgery (SSRF)

CVSS scores (2)

Score	Severity	Version	Source	Vector
8.6	HIGH	3.1	cna	CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:N/A:N
6.9	MEDIUM	4.0	cna	CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:N/VI:N/VA:N/SC:H/SI:N/SA:N

References (2)