Fill-mask

Inference Providers

Mask filling is the task of predicting the right word (token to be precise) in the middle of a sequence.

For more details about the fill-mask task, check out its dedicated page! You will find examples and related materials.

Recommended models

FacebookAI/xlm-roberta-base: A multilingual model trained on 100 languages.

Explore all available models and find the one that suits you best here.

Using the API

Language

Client

Provider

Settings

import os
from huggingface_hub import InferenceClient

client = InferenceClient(
    provider="hf-inference",
    api_key=os.environ["HF_TOKEN"],
)

result = client.fill_mask(
    "The answer to the universe is undefined.",
    model="google-bert/bert-base-uncased",
)

API specification

Request

Headers
authorization	string	Authentication header in the form `'Bearer: hf_**'` when `hf_**` is a personal user access token with “Inference Providers” permission. You can generate one from your settings page.

Payload
inputs*	string	The text with masked tokens
parameters	object
top_k	integer	When passed, overrides the number of predictions to return.
targets	string[]	When passed, the model will limit the scores to the passed targets instead of looking up in the whole vocabulary. If the provided targets are not in the model vocab, they will be tokenized and the first resulting token will be used (with a warning, and that might be slower).

Response

Body
(array)	object[]	Output is an array of objects.
sequence	string	The corresponding input with the mask token prediction.
score	number	The corresponding probability
token	integer	The predicted token id (to replace the masked one).
token_str	string	The predicted token (to replace the masked one).

Update on GitHub