- Home
- Resources
- Integrations
- Hugging Face
Hugging Face Make integrationAutomate Hugging Face with Make.
What if the text piling up in your other tools could be summarized, translated or sorted without a single copy and paste? The Hugging Face Make integration gives you 7 modules: six actions and one search. This page helps you connect your account and build a first working scenario.
Verified Trustpilot reviews · AI, automation & growth agency
What can the Hugging Face Make integration do for you?
The Hugging Face Make integration lets a Make scenario, the automation you build as a chain of bricks, send text to Hugging Face and get back an answer, a summary, a translation, a label or an image. Each brick is called a module, and none of them needs a line of code.
Shorter reading for your team. A long text arrives from another app, Create a Summary turns it into a few lines, and the next module posts that digest wherever your team reads it.
Content that speaks every language of your customers. A product description or a support reply goes through Create a Translation before it is sent, so the same scenario serves several markets.
A first sort on incoming messages. Create a Text Classification attaches a category to each message, and a filter in Make routes it to the right person. For open questions, Create a Chat Completion (Prompt) and Create an Answer write the reply, and Generate an Image draws a visual from a text prompt.
Hugging Face has no trigger in Make: it never starts a scenario by itself. Something else does, another app or a schedule, and Hugging Face works in the middle of the chain. When a listed module does not cover your need, Make an API Call reaches the rest of the API. The same app also exists on n8n: see the Hugging Face n8n integration and our n8n vs Make comparison, or start with the Make training if the editor is new to you.
How do you connect your Hugging Face account to Make?
- 01
Add a Hugging Face module
In your scenario, click the + and search for Hugging Face, then pick any of its modules. On the connection line, click Create a connection. A connection is your account linked to Make once and reused by every Hugging Face module.
- 02
Name it and authorize Make
Give the connection a name if you like, such as your team or project, so you can spot it later. Then authorize Make on the Hugging Face page that opens, or paste the key Hugging Face gives you, depending on what the window asks for.
- 03
Save and reuse it
Click Save. The connection now appears in the list of every Hugging Face module you add, in this scenario and the next ones, so the authorization happens only once per account.
Your first scenario with Hugging Face and Make
GoalWhen a new text lands in another app, Make summarizes it with Hugging Face and translates the summary.
- 01
Create the scenario
Open the Scenarios page, create a new scenario and click the + to add the first module. Choose the app where your texts arrive, for example a form tool or a spreadsheet, and its module that picks up new items.
- 02
Add Create a Summary
Click the + on the right of that first module, search for Hugging Face and pick Create a Summary. Select your connection, or click Create a connection if this is your first Hugging Face module.
- 03
Map the text to summarize
In the module fields, map the text coming from the first module. Each item flowing between modules is a bundle: one message, one row, one entry. Make then summarizes each bundle separately.
- 04
Chain Create a Translation
Add Create a Translation after the summary and map the summary text into it. The output is a short version of the original text, in the language your readers expect.
- 05
Test with Run once
Click Run once. Make runs the scenario one time, and a bubble on each module shows the bundles it received. Open them to check the summary and the translation before going further.
- 06
Schedule and switch it on
Click the clock on the first module to set the schedule, every 15 minutes by default, then switch the scenario on. Each module run on a bundle counts as one operation on your plan.
What each Hugging Face module does in Make
Hugging Face gives you 7 modules. For each one: what it does for you, when to reach for it, and what to watch out for.
Create a Chat Completion (Prompt)
ActionIn Make“Generates a response given a list of messages in a conversational context.”
This module sends a conversation, a list of messages, to a Hugging Face model and gets back the next reply. You obtain a written response that takes the earlier turns into account, not just the last line.
Create an Answer
ActionIn Make“Creates an answer for a given question.”
Hand it one question and it writes one answer. It fits cases where each request stands alone and no conversation history matters, which keeps the mapping short in the editor.
Create a Summary
ActionIn Make“Summarizes longer text into shorter text.”
Long text goes in, a shorter version comes out. Your readers get the gist of a report, an email thread or a transcript without opening the original document.
Create a Translation
ActionIn Make“Creates a translation of the text.”
You get the same text in another language, ready to be sent or stored by the next module. It lets one scenario feed several markets instead of maintaining one copy per language by hand.
Generate an Image
ActionIn Make“Generates an image based on a given text prompt.”
Describe a picture in words and this module produces the image. It turns a line of text coming from another app into a visual you can pass along to the rest of the scenario.
Make an API Call
ActionIn Make“Performs an arbitrary authorized API call.”
This one opens the rest of the Hugging Face API to your scenario, with the same connection as the other modules. It covers what the six ready-made modules leave out.
Create a Text Classification
SearchIn Make“Creates a text classification.”
It reads a text and tells you which category it belongs to. Unlike the other modules, it is a search: it can return zero, one or several results, each as its own bundle.
Need help automating Hugging Face with Make?
A person reads every message.