digline

Overview

digline is a Python tool for regression testing LLM applications. It stores baseline outputs in your repository, allowing you to compare new model responses against known-good results without relying on external servers.

Best for

Best for
Teams wanting self-hosted regression testing for LLM apps

Use cases

• Catching regressions in LLM responses
• Validating model updates against stored baselines
• Integrating regression checks into CI pipelines

Notes

digline is a Python tool for regression testing LLM applications. It stores baseline outputs in your repository, allowing you to compare new model responses against known-good results without relying on external servers.

7 stars on GitHub. Last updated 2026-10-01. Licensed Apache-2.0.

Use cases

• Catching regressions in LLM responses
• Validating model updates against stored baselines
• Integrating regression checks into CI pipelines

Pros

• Baselines live in your repo, no external service needed
• Simple Python integration
• Open source with community support

Cons

• Small community (7 stars) may mean limited documentation
• Requires manual baseline management
• Python-only, not language-agnostic

Indexed from awesome-llmops and enriched against its public facts.