---
title: "Obtain responsible prompt recommendations from a model running locally"
method: GET
path: "/recommend_local"
tags: ["recommend_local"]
---

# Obtain responsible prompt recommendations from a model running locally

`GET /recommend_local`

Given an input prompt with one or more sentences, returns up to five recommendations of sentences and its respective positive value representations to be added at the end of the input prompt, and up to five recommendations for removal of potentially harmful sentences in the given input prompt.

## Query parameters

- `prompt` string, required

## Headers

- `add_lower_threshold` string
- `add_upper_threshold` string
- `remove_lower_threshold` string
- `remove_upper_threshold` string
- `model_id` string, required
- `model_path` string, required

## Response `200`

successful recommendation

## Other responses

- `404` — invalid prompt

## Changes

- **2024-10-24** `7c6904b5a7a0` — 1 info
  - endpoint added
- **2024-08-09** `c9a4b3a38ef7` — 1 breaking
  - api path removed without deprecation

[Change history](https://skmtc.dev/ibm/apis/responsible-prompt-recommendation-api/changes/recommend_local/get.md)

---

[API](https://skmtc.dev/ibm/apis/responsible-prompt-recommendation-api.md) · [All operations](https://skmtc.dev/ibm/apis/responsible-prompt-recommendation-api/llms.txt) · [OpenAPI document](https://skmtc-service-production.skmtc.workers.dev/v1/apis/ibm/responsible-prompt-recommendation-api/revisions/289f88c3eca1/schema)