features

A merge gate built for non-deterministic software.

Drift-CI scores your LLM outputs against committed baselines and fails the build when behaviour drifts past your threshold — with everything you need to triage it.

The hosted dashboard

Drift-CI project dashboard listing recent runs with pass rates and scores

Run history — Free

Drift-CI issue page for a regressed case, showing its timeline, an assignee and team comments

Regression issues — Team

Score trend chart dipping and recovering over 30 days

Analytics — Team

providers

Every provider, one suite

Anthropic, OpenAI, Azure, Bedrock, Gemini, Vertex AI, and Ollama. Swap models and compare behaviour without rewriting a single case.

evaluators

An evaluator for every check

exact-match, json-schema, cosine-similarity, LLM-judge, refusal-detection, rubric-checklist, and safety-classifier — chained per case.

baselines

Baselines as code, not hidden state

One git-committed JSON file per case, with secrets redacted before anything touches disk. Review drift as a diff, in your PR.

ci-gate

A real merge gate

A regression exits 1 and fails the build. A transient-error storm exits 2 — never mistaken for a regression. Deterministic by design.

timelines

Drift timelines that catch slow rot

Visualize how scores evolve across runs, so you catch latent degradation that one-off evals miss.

alerts

Alerts where your team lives

Alert rules ship in the self-hosted dashboard today. Hosted Slack and email alerts are coming soon.

works with your stack

AnthropicOpenAIAzure OpenAIBedrockGeminiVertex AIOllama

drift_timeline

Score stability · last 12 runs

drift
run #112baseline updaterun #124

visibility

From a red X to a root cause.

The hosted dashboard keeps full run history and run and case diffs, and on Team adds cost analytics, so you can answer “what changed, and when?” in seconds — not by grepping CI logs.

Get started free