---
title: "✨ Compare correctness transitions between inference runs"
method: GET
path: "/api/inference-runs/regressions/"
tags: ["Prompts"]
---

# ✨ Compare correctness transitions between inference runs

`GET /api/inference-runs/regressions/`

Compare correctness transitions between two inference runs.

Classifies each task as regression, improvement, stable_correct,
stable_incorrect, new, or removed based on prediction scores.

## Response `200`

- RegressionsResponse
  - `improvements` unknown[], required
    - unknown
  - `new` unknown[], required
    - unknown
  - `regressions` unknown[], required
    - unknown
  - `removed` unknown[], required
    - unknown
  - `run_a` integer, required
  - `run_b` integer, required
  - `stable_correct` unknown[], required
    - unknown
  - `stable_incorrect` unknown[], required
    - unknown
  - `summary` RegressionSummary, required
    - `improvements` integer, required
    - `new` integer, required
    - `regressions` integer, required
    - `removed` integer, required
    - `stable_correct` integer, required
    - `stable_incorrect` integer, required

---

[API](https://skmtc.net/humansignal/apis/label-studio-api.md) · [All operations](https://skmtc.net/humansignal/apis/label-studio-api/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/humansignal/label-studio-api/versions/1b113b8df950/schema)