---
id: CVE-2026-10803
aliases:
  - GHSA-5qmp-p3c4-72qj
  - BIT-mlflow-2026-10803
  - PYSEC-2026-195
title: >-
  MLflow: Deterministic sampling in dataset digest enables predictable
  collisions
summary: >-
  MLflow: Deterministic sampling in dataset digest enables predictable
  collisions
severity: low
cvss: 3.6
cvssVector: 'CVSS:3.1/AV:L/AC:H/PR:L/UI:N/S:U/C:N/I:L/A:L'
vendor: mlflow
product: mlflow
ecosystem: pip
affected:
  - mlflow < 3.10.1
patched:
  - mlflow 3.10.1
published: '2026-06-04'
updated: '2026-07-15'
source: OSV
sourceUrl: 'https://osv.dev/vulnerability/GHSA-5qmp-p3c4-72qj'
references:
  - url: 'https://nvd.nist.gov/vuln/detail/CVE-2026-10803'
  - url: 'https://github.com/mlflow/mlflow/issues/22419'
  - url: 'https://github.com/mlflow/mlflow/pull/22420'
  - url: 'https://github.com/mlflow/mlflow'
  - url: >-
      https://github.com/pypa/advisory-database/tree/main/vulns/mlflow/PYSEC-2026-195.yaml
  - url: 'https://vuldb.com/cve/CVE-2026-10803'
  - url: 'https://vuldb.com/submit/831462'
  - url: 'https://vuldb.com/vuln/368252'
  - url: 'https://vuldb.com/vuln/368252/cti'
tags:
  - osv
  - pip
epss: 0.00103
epssPercentile: 0.00906
ingestedAt: '2026-07-16T18:59:41.471Z'
---

## Overview

A flaw has been found in MLflow up to 3.10.0. This issue affects the function mlflow.data.digest_utils of the file mlflow/data/digest_utils.py of the component Dataset Digest Computation. This manipulation causes use of weak hash. It is possible to launch the attack on the local host. The attack is considered to have high complexity. The exploitability is assessed as difficult. The exploit has been published and may be used. The project was informed of the problem early through a pull request but has not reacted yet.

## Affected packages

- `mlflow < 3.10.1`

## Remediation

Upgrade to a patched release:

- `mlflow 3.10.1`