Developer MaxiDonkey has released DelphiGemini, a native Google Gemini API wrapper for Delphi designed to bring advanced multimodal AI capabilities directly into Delphi and FireMonkey applications.
DelphiGemini enables developers to integrate Gemini models for text generation, conversational AI, embeddings, vector search, structured outputs, image and video generation, and agent-based workflows entirely from Delphi code. The wrapper provides support for stateless generation as well as persistent AI interactions with streaming and asynchronous workflows.
The library includes support for:
- Interactive chat and conversational AI
- Multimodal input including text, image, audio, video, and PDF
- Embeddings and semantic vector search
- Structured outputs and function calling
- Streaming responses and async workflows
- Batch processing and caching
- Grounding with Google Search and Maps
- Image and video generation with Gemini models
- Agent workflows and persistent interactions
Firemonkey is built for Windows, macOS, Linux, Android, and iOS. DelphiGemini probably supports cross-platform development with Delphi and FireMonkey while exposing Gemini’s latest AI capabilities through a native Delphi API.
The project aims to provide Delphi developers with a production-ready alternative to official SDKs that are not currently available for Delphi, while maintaining broad access to Gemini platform features without abstraction limitations.
Learn more about DelphiGemini and download the full source code on GitHub.
