ChatBot Portal for Multi-Modal LLM APIs
The GPTPortal is a robust web application, effortlessly combining advanced technologies to provide an interactive chatbot experience. Centered around the prowess of OpenAI’s GPT-4 API, this project unlocks new frontiers with its vision, hearing, and speaking abilities, alongside capabilities like image generation, file uploads, and customizable instructions in system prompts.
Key Features
1. Interactive Interface
Users can comfortably chat through a sleek text message system. The addition of voice conversations via Whisper, spanning speech transcription and spoken responses, elevates this interaction. Image uploads further enrich conversations, adding a layer of visual context.
2. Extensive API Integration
The portal seamlessly integrates with seven AI API providers, including Anthropic Claude and Mistral AI, alongside Google Gemini, using free API access.
3. User-Friendly Design
Security is maintained with basic authentication, and the UI is designed for simplicity. Ease-of-use features like copy-to-clipboard and markdown rendering create a seamless user experience.
4. Advanced Chat Functions
Conversations can be saved in HTML format, and the portal features an integrated shutdown via a simple command: “Bye!”. For deeper interaction, image generation with DALL·E 3 is easily accessible.
5. Custom Environment for Developers
Through Assistants Mode, users can engage within a native code environment, leveraging OpenAI's Beta API to reuse files and instructions. Automatic Python execution is facilitated in a Jupyter-like environment, supporting retrieval-augmented generation from uploaded files.
6. Extensive Model Support
The project supports a slew of high-performance models. With a model selector offering GPT-4, Gemini, Claude Opus-Instant, and Mistral models, users can explore a variety of language processing capabilities.
Additional Features
- Automatic Token Cost Calculations: Stay informed with real-time updates on token costs and summaries.
- Editable Custom Instructions: Modify system instructions conveniently through a frontend interface.
- Keyboard Shortcuts & Prompt Templates: Enhance usability with shortcuts and pre-defined prompt templates for ease of use.
- Conversation History: Efficiently track interactions with optimized context windows.
Pricing
The project offers a detailed model pricing table, differentiating costs associated with various models. Some models like the Gemini series and LLaMA-3 come at no cost for token usage, demonstrating a cost-effective option for users.
Comprehensive Documentation and Updates
The project provides an expansive array of documentation and regular updates, ensuring users have access to the latest features and improvements. A model pricing table gives clear insight into costs related to each supported model, encouraging transparency and informed decision-making.
Conclusion
The GPTPortal is more than just a chat interface; it's a pioneering platform integrating multiple AI systems under one umbrella. Offering advanced capabilities, customization options, and extensive model support, this project is set to redefine how users interact with AI across various modalities. Whether for casual users or developers, the GPTPortal provides a versatile and powerful solution for AI-based interaction.