TheoKanning/openai-javaArchived
Openai Java
openai-java ist eine Java-Client-Bibliothek und ein SDK, das entwickelt wurde, um OpenAI-Dienste in Java-Anwendungen zu integrieren. Es fungiert als programmatischer Wrapper für Endpunkte von Large Language Models und ermöglicht die Implementierung generativer KI-Funktionen wie Chat-Completions, Text-Embeddings und die Orchestrierung von KI-Agenten.
Die Bibliothek unterstützt eine breite Palette multimodaler Aufgaben, einschließlich der Generierung und Bearbeitung von Bildern sowie der Audiotranskription und -übersetzung. Sie bietet spezialisierte Tools für das Modelltraining und Fine-Tuning, wodurch Nutzer Datensätze hochladen können, um die Modellleistung in spezifischen Domänen zu verbessern. Zudem enthält sie Funktionen für die Ausführung benutzerdefinierter Funktionen und Content-Moderation, um Ein- und Ausgaben anhand von Sicherheitsrichtlinien zu filtern.
Das SDK verwendet ein Builder-Pattern für die Request-Konstruktion und mappt API-Ressourcen auf stark typisierte Java-Klassen. Es unterstützt sowohl synchrone Request-Response-Zyklen als auch die inkrementelle Datenbereitstellung mittels Server-Sent-Events für Echtzeit-Response-Streaming.
Features
- AI Service Integrations - Serves as a Java client library for connecting local applications to external OpenAI model endpoints.
- Generative AI SDKs - Provides a comprehensive Java toolkit for implementing AI agents, custom function calling, and model fine-tuning.
- Conversational AI Assistants - Facilitates the development of context-aware chatbots and virtual assistants within Java applications.
- Conversational AI Models - Handles multi-turn dialogues by exchanging formatted message histories between the user and the model.
- External LLM API Wrappers - Wraps external large language model APIs into a programmatic Java interface for structured request and response handling.
- OpenAI API Clients - Implements a client for sending prompts and receiving responses from OpenAI generative AI models.
- Text Completion Engines - Generates human-like text responses based on prompts using large language models.
- REST API Integrations - Implements integration logic to communicate with remote AI services via RESTful HTTP endpoints.
- AI Response Streams - Supports receiving AI-generated text incrementally via server-sent events to reduce perceived latency.
- AI Agent Orchestration - Provides frameworks for coordinating agents that combine custom instructions, tools, and retrieval.
- AI Agent State Coordination - Provides tools for managing the execution state and tool interactions of persistent AI assistants.
- Audio Transcriptions - Converts spoken audio into written text and supports audio-based translation.
- Custom Model Training - Manages the process of fine-tuning generative models on specialized datasets.
- Multimodal Generative AI - Integrates a pipeline for generating images and transcribing audio from text-based prompts.
- Multimodal AI Toolkits - Enables automated speech transcription and AI-powered image generation through a single interface.
- Natural Language Processing - Provides capabilities for generating numerical embeddings and performing safety moderation on natural language text.
- Text Embedding Generators - Converts written input into numerical vector embeddings for semantic similarity searches.
- Real-time Tool Execution - Enables the AI to trigger external API actions and execute custom logic during live interactions.
- Image Generation And Editing - Creates new images from text descriptions and modifies existing images via generative prompts.
- Model Fine-Tuning - Supports uploading datasets and training custom model versions to optimize performance for specific domains.
- Server-Sent Events - Processes incremental HTTP responses as a stream to handle real-time text generation.
- Language SDKs - Java client library with support for advanced models.
- Microservices - Library for interacting with OpenAI GPT APIs.
- Programming Libraries - Java client library for the OpenAI API.
