• Documentation
    • About ​ValidMind
    • Get Started
    • Guides
    • Support
    • Releases

    • ValidMind Library
    • Python API
    • Public REST API

    • Training Courses
  • Log In
  1. 2.10.0 Release notes
  • Installing ​ValidMind
  • Release notes
  • ​ValidMind systems access
  • Troubleshooting and support

  • Deployment
  • 1 — Prepare your environment
  • 2 — Pull the container images
  • 3 — Pull the Helm chart
  • 4 — Configure your Helm chart
  • 5 — Configure LiteLLM
  • 6 — Install or upgrade via Helm chart
  • 7 — Complete the initial setup
  • 8 — Verify your installation

  • Authentication
  • Configuring authentication
  • Configure local authentication
  • Configure OIDC integration
  • Authenticate with Entra
    • Configure frontend environment variables for Entra
    • Entra manifest JWT configuration
    • Configure single sign-on (SSO)
  • Supporting multiple OIDC providers

  • Administering
  • Operating ​ValidMind
  • Migrate to vmconfig.toml (26.02)
  • View ValidMind configuration
  • Configure AI model settings
  • Monitoring your deployment
  • Configuring email services
    • Configure email provider
    • Customize email templates

  • Reference
  • ​ValidMind Admin REST API
  • ValidMind configuration

  • Security advisories
  • VM-SA-2026-01
  • VM-SA-2026-02

On this page

  • Enhancements
    • Add new deepeval tests in library
  • Edit this page
  • Report an issue

2.10.0 Release notes

validmind-library
2.10.0
patch release
Published

October 17, 2025

This release includes new deepeval tests in the library.

Enhancements

Add new deepeval tests in library

This update enhances the integration between ValidMind and DeepEval by introducing primary scorers within a dedicated deepeval namespace. You can now evaluate large language model (LLM) outputs using standardized metrics. Primary scorers are essential tools for evaluating LLM outputs. We have added several DeepEval-based LLM scorers, including:

  • Hallucination
  • Contextual precision/recall
  • Summarization
  • Task completion

Additionally, a demo notebook is available to guide you through end-to-end usage. Improvements have also been made to plots, such as boxplots, and examples.

DeepEval Integration with ValidMind

  • ValidMind Logo
    ©
    Copyright 2026 ValidMind Inc.
    All Rights Reserved.
    Cookie preferences
    Legal
  • Get started
    • Development
    • Validation
    • Setup & admin
  • Guides
    • Access
    • Configuration
    • Integrations
    • Workflows
    • Inventory
    • Risk tiering
    • Documents & templates
    • Documentation
    • Validation
    • Reporting & auditing
    • Monitoring
    • Attestation
  • ValidMind Library
    • Quickstarts
    • Development tutorial
    • Validation tutorial
    • Run tests & test suites
    • Use library features
    • Code samples
    • Python API
    • Public REST API
  • Training
    • Learning paths
    • Courses
    • Videos
  • Support
    • Troubleshooting
    • FAQ
    • Get help
  • Edit this page
  • Report an issue
  • Community
    • GitHub
    • LinkedIn
    • Events
    • Blog