Hugging Face Text - AI Text Processing with 100K+ NLP Models

Process text with the world's largest collection of NLP models. Text generation, summarization, translation, sentiment analysis, NER, Q&A, embeddings, paraphrase, and zero-shot classification. Lightweight actor focused on text-only tasks. BYOK with your Hugging Face API token.

Features

• Text Generation - Generate content with Llama 2, Mistral, Falcon
• Chat Completion - Conversational AI with system prompts
• Summarization - Condense documents with BART, T5
• Translation - 100+ language pairs with Helsinki-NLP
• Sentiment Analysis - Classify emotions and sentiment
• Named Entity Recognition - Extract people, places, organizations
• Question Answering - Extract answers from context
• Fill-Mask - Complete sentences with BERT
• Embeddings - Generate semantic vectors
• Zero-Shot Classification - Classify without training data
• Text Similarity - Compare text semantically
• Paraphrase - Rewrite text in different ways
• Demo Mode - Test with sample data before going live

Who Should Use This Actor?

Content Marketers

Generate blog posts, social content, and copy. Summarize research. Repurpose content across formats.

SEO Specialists

Generate meta descriptions. Expand keyword content. Translate for international SEO.

Customer Support Teams

Analyze ticket sentiment. Extract key entities. Generate response templates.

Data Analysts

Extract entities from text data. Classify documents. Generate embeddings for ML pipelines.

Researchers

Summarize papers. Translate sources. Analyze sentiment in datasets.

E-commerce Teams

Generate product descriptions. Translate listings. Analyze reviews.

Quick Start

Demo Mode (Free Test)

{
  "task": "text_generation",
  "prompt": "Write a product description for wireless earbuds",
  "demoMode": true
}

Text Generation

{
  "task": "text_generation",
  "apiToken": "hf_your_token_here",
  "prompt": "Write a compelling meta description for a page about local SEO services",
  "maxTokens": 160,
  "temperature": 0.7,
  "demoMode": false
}

Chat Completion

{
  "task": "chat_completion",
  "apiToken": "hf_your_token_here",
  "model": "mistralai/Mistral-7B-Instruct-v0.1",
  "systemPrompt": "You are a helpful SEO expert. Provide concise, actionable advice.",
  "prompt": "How do I improve my local search rankings?",
  "maxTokens": 300,
  "demoMode": false
}

Summarization

{
  "task": "summarization",
  "apiToken": "hf_your_token_here",
  "text": "Your long article or document text goes here...",
  "maxTokens": 150,
  "minTokens": 50,
  "demoMode": false
}

Sentiment Analysis

{
  "task": "sentiment_analysis",
  "apiToken": "hf_your_token_here",
  "text": "This product exceeded my expectations! Amazing quality and fast shipping.",
  "demoMode": false
}

Named Entity Recognition

{
  "task": "ner",
  "apiToken": "hf_your_token_here",
  "text": "Tim Cook announced that Apple will open a new store in San Francisco next month.",
  "demoMode": false
}

Translation

{
  "task": "translation",
  "apiToken": "hf_your_token_here",
  "text": "Our services are available in all major cities.",
  "sourceLanguage": "en",
  "targetLanguage": "fr",
  "demoMode": false
}

Embeddings

{
  "task": "embeddings",
  "apiToken": "hf_your_token_here",
  "text": "Local SEO services for small businesses",
  "demoMode": false
}

Batch Embeddings

{
  "task": "embeddings",
  "apiToken": "hf_your_token_here",
  "texts": "[\"SEO services\", \"Digital marketing\", \"Local business optimization\"]",
  "demoMode": false
}

Zero-Shot Classification

{
  "task": "zero_shot_classification",
  "apiToken": "hf_your_token_here",
  "text": "The new iPhone features a revolutionary camera system",
  "candidateLabels": "technology,business,sports,entertainment",
  "demoMode": false
}

Text Similarity

{
  "task": "text_similarity",
  "apiToken": "hf_your_token_here",
  "text": "SEO optimization services",
  "compareText": "Search engine marketing",
  "demoMode": false
}

Paraphrase

{
  "task": "paraphrase",
  "apiToken": "hf_your_token_here",
  "text": "Local SEO helps small businesses get found online",
  "maxTokens": 100,
  "demoMode": false
}

Input Parameters

Parameter	Type	Default	Description
task	string	required	Task to perform (see task list)
apiToken	string	-	Your Hugging Face API token
model	string	task default	Specific model to use
text	string	-	Input text for processing
texts	string	-	JSON array of texts (for batch embeddings)
prompt	string	-	Input prompt for generation
systemPrompt	string	-	System prompt for chat completion
context	string	-	Context for Q&A
question	string	-	Question for Q&A
compareText	string	-	Second text for similarity
sourceLanguage	string	"en"	Source language code
targetLanguage	string	"es"	Target language code
candidateLabels	string	-	Comma-separated classification labels
maxTokens	number	256	Max tokens to generate
minTokens	number	30	Min tokens for summarization
temperature	number	0.7	Randomness (0.0-1.0)
topP	number	0.9	Nucleus sampling
topK	number	50	Top-k sampling
repetitionPenalty	number	1.0	Reduce repetition
doSample	boolean	true	Enable sampling
waitForModel	boolean	true	Wait for model to load
webhookUrl	string	-	Webhook URL for results
demoMode	boolean	true	Return sample data

Available Tasks

Task	Description	Default Model
text_generation	Generate text	Mistral-7B-Instruct
chat_completion	Conversational AI	Mistral-7B-Instruct
summarization	Summarize text	BART-large-CNN
translation	Translate text	Helsinki-NLP OPUS
sentiment_analysis	Analyze sentiment	DistilBERT-SST2
ner	Named entity recognition	BERT-NER
question_answering	Answer from context	RoBERTa-SQuAD2
fill_mask	Complete sentences	BERT-base
embeddings	Vector embeddings	all-MiniLM-L6-v2
zero_shot_classification	Classify without training	BART-large-MNLI
text_similarity	Compare texts	all-MiniLM-L6-v2
paraphrase	Rewrite text	Pegasus-paraphrase

Output Format

Text Generation

{
  "success": true,
  "model": "mistralai/Mistral-7B-Instruct-v0.1",
  "generatedText": "Boost your local visibility with our proven SEO strategies..."
}

Sentiment Analysis

{
  "success": true,
  "model": "distilbert-base-uncased-finetuned-sst-2-english",
  "sentiment": [
    {"label": "POSITIVE", "score": 0.89},
    {"label": "NEGATIVE", "score": 0.11}
  ]
}

Named Entity Recognition

{
  "success": true,
  "model": "dslim/bert-base-NER",
  "entities": [
    {"entity": "B-PER", "word": "Tim", "score": 0.99, "start": 0, "end": 3},
    {"entity": "I-PER", "word": "Cook", "score": 0.98, "start": 4, "end": 8},
    {"entity": "B-ORG", "word": "Apple", "score": 0.97, "start": 24, "end": 29},
    {"entity": "B-LOC", "word": "San", "score": 0.95, "start": 56, "end": 59},
    {"entity": "I-LOC", "word": "Francisco", "score": 0.96, "start": 60, "end": 69}
  ]
}

Zero-Shot Classification

{
  "success": true,
  "model": "facebook/bart-large-mnli",
  "labels": [
    {"label": "technology", "score": 0.85},
    {"label": "business", "score": 0.12},
    {"label": "entertainment", "score": 0.02},
    {"label": "sports", "score": 0.01}
  ]
}

Embeddings

{
  "success": true,
  "model": "sentence-transformers/all-MiniLM-L6-v2",
  "embedding": [0.023, -0.156, 0.089, ...],
  "dimensions": 384
}

Text Similarity

{
  "success": true,
  "model": "sentence-transformers/all-MiniLM-L6-v2",
  "similarity": 0.85,
  "text1": "SEO optimization services",
  "text2": "Search engine marketing"
}

Paraphrase

{
  "success": true,
  "model": "tuner007/pegasus_paraphrase",
  "paraphrases": [
    "Small businesses can improve their online presence with local SEO",
    "Getting found online is easier for small businesses with local SEO",
    "Local search optimization benefits small business visibility"
  ]
}

Pricing (Pay-Per-Event)

Event	Description	Price
text_processed	Per text processing task	$0.01

Example costs:

• 100 sentiment analyses: 100 × $0.01 = $1.00
• 50 translations: 50 × $0.01 = $0.50
• 200 embeddings: 200 × $0.01 = $2.00
• Demo mode: $0.00

Common Scenarios

Scenario 1: SEO Content Generation

{
  "task": "text_generation",
  "apiToken": "hf_your_token",
  "prompt": "Write 5 unique meta descriptions for a local plumbing company in Miami",
  "maxTokens": 500,
  "temperature": 0.8,
  "demoMode": false
}

Scenario 2: Review Sentiment Analysis

{
  "task": "sentiment_analysis",
  "apiToken": "hf_your_token",
  "text": "The service was okay but took longer than expected. Staff were friendly though.",
  "demoMode": false
}

Scenario 3: Multilingual SEO

{
  "task": "translation",
  "apiToken": "hf_your_token",
  "text": "Professional plumbing services available 24/7. Call now for a free quote.",
  "sourceLanguage": "en",
  "targetLanguage": "es",
  "demoMode": false
}

Scenario 4: Content Categorization

{
  "task": "zero_shot_classification",
  "apiToken": "hf_your_token",
  "text": "New research shows that regular exercise improves mental health",
  "candidateLabels": "health,fitness,science,lifestyle,news",
  "demoMode": false
}

Webhook & Automation Integration

Zapier / Make.com / n8n

1. Create a webhook trigger
2. Copy the URL to webhookUrl
3. Process text results in your workflow

Popular automations:

• Generated content -> WordPress/CMS
• Sentiment scores -> CRM tagging
• Translations -> Multilingual site updates
• NER extraction -> Contact enrichment

Hugging Face AI Suite

Actor	Best For
Hugging Face Master	All-in-one (text + image + audio)
Hugging Face Text	Text-only processing (lightweight)
Hugging Face Image	Image processing
Hugging Face Audio	Audio processing
Hugging Face Hub	Model discovery

FAQ

Q: What's the max text length?

A: Most models support 512-4096 tokens. For longer texts, chunk and process separately.

Q: How do I choose a model?

A: Start with defaults - optimized for quality/speed. Check Hugging Face for task-specific leaderboards.

Q: Can I use multilingual models?

A: Yes! Many models support 100+ languages. Check model cards for language support.

Q: Are batch requests supported?

A: Yes for embeddings - pass a JSON array in texts. Other tasks process one at a time.

Common Problems & Solutions

"Model is loading"

• Large models need warm-up time
• Set waitForModel: true (default)
• Use smaller models for faster responses

"Text too long"

• Split into smaller chunks
• Use summarization first, then process

"Demo data showing"

• Set demoMode: false
• Provide your Hugging Face API token

---

Built by John Rippy | Actor Arsenal