Inquiry About AI Tools and Integration

[Shukino20001015]

The AI tools on the market are iterating quite rapidly, with new products constantly emerging. I would like to know how their update and maintenance processes work, and whether they support custom extensions for basic programming personnel. I have only looked through your documentation, and I resonate with your ideas, but I need to consider the practicality of integrating the software and whether the overall user experience will become more cumbersome due to a decentralized app. For example, if I have my own locally deployed workflow, will I be able to conveniently integrate it? Essentially, I'm concerned about ensuring the ease of use and extensibility of the software after the deployment of decentralized computing power. I want to share it with my team and friends who may not have development skills but need to use these tools. Furthermore, I will also want to continuously integrate new products.

[fnalways]

Hi @Shukino20001015 - thank you for your thoughtful questions. We completely understand your concerns regarding the integration, extensibility, maintenance and usability of AI applications.

Instead of focusing on specific details, we would like to offer a broader perspecitve on Olares' roadmap in terms of AI. Generally, Olares has three primary goals:

• Short-term (user-focused): Make it easy for users to deploy open-source AI projects on their servers while ensure secure access anytime, anywhere.
• Mid-term (developer-focused): Enable developers to integrate user data as well as various AI applications.
• Long-term (AI Agent-focused): Facilitate collaboration between AI agents across different users online.

Currently, we are in the mid-term phase - to enable developers to flexibly combine different functional components within Olares. The key areas now we are addressing:

• Knowledge organization: Collect and index documents from SaaS systems (e.g., Google Drive), devices (via built-in Seafile), or the web (via the Wise app).
• Knowledge indexing: Creates inverted indexes for full-text search and vector databases for RAG systems.
• GPU Management: Optimize GPU usage across various AI applications.
• Model integration: Support both local models (e.g., LLMs, Whisper, Stable Diffusion) and remote APIs (e.g., OpenAI).
• Agent and workflow creation: Enable automation using tools like Dify and LlamaIndex, available to connect workflows via APIs.
• Device integration: Ensure a consistent experience across mobile, PC, and browser plugins while utilizing the power of home hardware for processing tasks.

For a practical example, we recommend reviewing our blog post. It explains how tools like Perplexica, Ollama, and SearXNG are combined to create an AI service. The main advantages that Olares offer are:

• Developers can easily migrate a third-party AI applications to Olares.
• Users can install these AI apps within a few clicks from the marketplace, and combine them flexibly.
• All apps are accessible through a single, secure domain, ensuring seamless access anytime anywhere.

Olares is commited to making AI approachable and accessible to all, including non-technical users. If you’d like to discuss a specific use case or have further questions, please share them here.