Promptify

Overview of Promptify

What it is

Promptify is a task-based NLP engine that provides Pydantic structured outputs, built-in evaluation, and uses LiteLLM as a universal LLM backend. It is described as 'scikit-learn for LLM-powered NLP' and is hosted on GitHub under the PromptsLab organization. The project is open source under the Apache-2.0 license.

Key features

• 2-3 lines of code for any NLP task, no training data required - Pydantic structured outputs for type-safe results - Supports any LLM provider via LiteLLM (OpenAI, Anthropic, Google, Ollama, Azure, and 100+ more)
• Built-in tasks: NER, Classification (binary/multiclass/multilabel), QA, Summarization, Relation Extraction, SQL Generation, and more - Custom tasks with any Pydantic schema - Few-shot examples and domain specialization (e.g., domain='medical')
• Batch processing with async concurrency and native async support (acall())
• Evaluation framework with precision, recall, F1, accuracy, exact match, ROUGE metrics - Safe parser with fallback JSON completion for providers without native structured outputs - Cost tracking via get_cost_summary()

Use cases

Promptify can be used for extracting named entities from medical text, classifying customer reviews, answering questions based on provided context, generating structured outputs from any text using custom Pydantic schemas, and processing multiple texts concurrently. It is suitable for developers and researchers who need to integrate LLM-powered NLP tasks into Python applications without training custom models.

Getting started

Requires Python 3.9+. Install with pip: bash pip install promptify Or install from the repository: bash pip install git+https://github.com/promptslab/Promptify.git For evaluation metrics support: bash pip install promptify[eval]

License and pricing

The project is licensed under Apache-2.0 and is open source. No pricing information is stated.

Based on the project's public README and repository metadata.