---
id: CVE-2026-115886
title: >-
  MLflow: SSRF on unauthenticated MLflow servers via third-party scorer kwargs
  in the scorer invoke, register, and GenAI evaluate endpoints
summary: >-
  MLflow is an open source AI engineering platform for agents, large language
  models, and machine learning models. From 3.12.0rc0 until 3.17.0, the scorer
  invoke, register, and GenAI evaluate endpoints, including unauthenticated POST
  /ajax…
severity: medium
cvss: 6.5
cvssVector: 'CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N'
cvssSource: cna
cwe:
  - CWE-470
  - CWE-918
vendor: mlflow
product: mlflow
affected:
  - 'mlflow >= 3.12.0rc0, < 3.17.0'
ssvc:
  exploitation: poc
  automatable: 'yes'
  technicalImpact: partial
  timestamp: '2026-10-29T18:17:37.566428Z'
exploitAvailable: true
published: '2026-10-23'
updated: '2026-10-29'
sourceUpdated: '2026-10-29T18:18:21.052Z'
source: CVEORG
sourceUrl: 'https://www.cve.org/CVERecord?id=CVE-2026-115886'
references:
  - url: 'https://github.com/mlflow/mlflow/security/advisories/GHSA-2wqp-q3c6-ccj5'
    label: 'https://github.com/mlflow/mlflow/security/advisories/GHSA-2wqp-q3c6-ccj5'
  - url: 'https://github.com/mlflow/mlflow/pull/25890'
    label: 'https://github.com/mlflow/mlflow/pull/25890'
  - url: >-
      https://github.com/mlflow/mlflow/commit/bac30b4148c4f40045b4d4db07ae60ab3838803f
    label: >-
      https://github.com/mlflow/mlflow/commit/bac30b4148c4f40045b4d4db07ae60ab3838803f
  - url: 'https://github.com/mlflow/mlflow/releases/tag/v3.17.0'
    label: 'https://github.com/mlflow/mlflow/releases/tag/v3.17.0'
  - url: 'https://github.com/mlflow/mlflow'
  - url: 'https://github.com/advisories/GHSA-2wqp-q3c6-ccj5'
tags:
  - cve.org
  - exploit-available
  - osv
  - pip
  - ghsa
epss: 0.00676
epssPercentile: 0.52344
exploits:
  nuclei:
    - CVE-2026-115886
  checkedAt: '2026-11-05T15:34:31.843Z'
aliases:
  - GHSA-2wqp-q3c6-ccj5
  - BIT-mlflow-2026-115886
ecosystem: pip
patched:
  - mlflow 3.17.0
ingestedAt: '2026-10-23T21:38:13.051Z'
---

## Overview

MLflow is an open source AI engineering platform for agents, large language models, and machine learning models. From 3.12.0rc0 until 3.17.0, the scorer invoke, register, and GenAI evaluate endpoints, including unauthenticated POST /ajax-api/3.0/mlflow/scorer/invoke, accept serialized third-party scorer data that Scorer.model_validate_json resolves using a module-path allowlist without constraining the selected class or its constructor arguments. When optional trulens and trulens-providers-litellm packages are installed, an attacker can select mlflow.genai.scorers.trulens, choose a model URI outside MLflow's native provider registry so execution uses the LiteLLM fallback, and supply api_base and api_key values that cause the server to send an outbound request to an attacker-chosen or internal host with attacker-controlled headers and body. The vulnerable deserialization occurs before a tracking-store lookup or job submission. This issue is fixed in version 3.17.0.

## Affected

- `mlflow >= 3.12.0rc0, < 3.17.0`

## Remediation

Refer to the linked advisories for vendor-supplied fixes and affected version ranges.

## Package advisory (CVE-2026-115886)

Affected packages:

- `mlflow >= 3.12.0rc0, < 3.17.0`

Patched in:

- `mlflow 3.17.0`

Source: https://osv.dev/vulnerability/GHSA-2wqp-q3c6-ccj5