CVE intelligence and bounded remediation

CVE-2025-61784 — LLaMA-Factory is a tuning library for large language models

High CVSS 8.1

LLaMA-Factory is a tuning library for large language models. Prior to version 0.9.4, a Server-Side Request Forgery (SSRF) vulnerability in the chat API allows any authenticated user to force the server to make arbitrary HTTP requests to internal and external networks. This can lead to the exposure of sensitive internal services, reconnaissance of the internal network, or interaction with third-party services. The same mechanism also allows for a Local File Inclusion (LFI) vulnerability, enabling users to read arbitrary files from the server's filesystem. The vulnerability exists in the `_process_request` function within `src/llamafactory/api/chat.py.` This function is responsible for processing incoming multimodal content, including images, videos, and audio provided via URLs. The function checks if the provided URL is a base64 data URI or a local file path (`os.path.isfile`). If neither is true, it falls back to treating the URL as a web URI and makes a direct HTTP GET request using `requests.get(url, stream=True).raw` without any validation or sanitization of the URL. Version 0.9.4 fixes the underlying issue.

Severity
High
CVSS
8.1 (3.1)
Published
2025-10-07
CISA KEV
Not currently listed
Ecosystem
software/application
Weaknesses
CWE-22, CWE-918

Affected products

  • hiyouga / llama-factory

Matched remediation archetype

Server-side request forgery and unintended proxying

This catalog composition supplies bounded fallback guidance. Explicitly reviewed curated workflows load with the complete record below.

Check exposure

  • Inventory server-side URL fetchers, webhooks, importers, previews, redirects, proxies, and protocol handlers reachable from untrusted input.
  • Map egress paths to internal services, metadata endpoints, loopback, private address space, and privileged control planes.
  • Review DNS resolution, redirect, proxy, credential-forwarding, and URL parsing behavior without requesting sensitive targets.

Remediate safely

  • Replace arbitrary destinations with named integrations or a strict allowlist of schemes, hosts, ports, and paths.
  • Resolve and validate every destination and redirect hop, then enforce egress policy independently of application checks.
  • Remove ambient credentials and sensitive headers from fetchers; apply response size, time, and content limits.

Authoritative sources

Complete CVE record and remediation plan

The detailed catalog view below loads this exact record, its source evidence, and the full seven-phase agentic change plan.