11.ai
11.ai is a voice-first AI assistant built on ElevenLabs Conversational AI that connects your voice to everyday workflows via the Model Context Protocol (MCP), enabling hands-free planning, research, project management, and team communication. By integrating out of the box with tools such as Perplexity for live web research, Linear for issue tracking, Slack for messaging, and Notion for knowledge management, and supporting custom MCP servers, 11.ai can interpret sequential voice commands, contextualize data, and take meaningful actions. It delivers real-time, low-latency interactions with multimodal support (voice and text), integrated retrieval-augmented generation, automatic language detection for seamless multilingual conversations, and enterprise-grade security (including HIPAA compliance).
Learn more
Gemini 2.5 Computer Use
Introducing the Gemini 2.5 Computer Use model, a specialized agent model built on top of Gemini 2.5 Pro’s visual reasoning capabilities, designed to interact directly with user interfaces (UIs). It is exposed via a new computer-use tool in the Gemini API, with inputs that include the user’s request, a screenshot of the UI environment, and a history of recent actions. The model generates function calls corresponding to UI actions like clicking, typing, or selecting, and may request user confirmation for higher-risk tasks. After each action is executed, a new screenshot and URL are fed back into the model to continue the loop until the task completes or is halted. It is optimized primarily for web browser control and shows promise for mobile UI interaction, though it is not yet suited for desktop OS-level control. In benchmarks across web and mobile control tasks, Gemini 2.5 Computer Use outperforms leading alternatives, delivering high accuracy at lower latency.
Learn more
Project Mariner
Project Mariner is a research prototype developed by Google DeepMind, built upon their advanced AI model, Gemini 2.0. It explores the future of human-agent interaction by automating tasks within a user's browser. Leveraging multimodal understanding, Project Mariner comprehends and reasons across various browser elements, including text, code, images, and forms. This enables it to navigate complex websites, automate repetitive tasks, and provide visual feedback to users. The system can interpret voice instructions and offers updates on task progress, ensuring users remain informed and in control. Additionally, Project Mariner can follow complex instructions by breaking them down into actionable steps, understanding relationships between web elements, and providing clear plans and actions to users. Currently, Project Mariner is in the testing phase with a select group of trusted users. Those interested in participating can join the waitlist for future testing opportunities.
Learn more
c/ua
c/ua is a platform that runs secure AI agents, optimized for Apple Silicon. It removes the need for virtual machine setup, enabling near-native macOS and Linux environments. Features include configurable VM resources, AI system integration, and automation via a computer-user interface. It supports multi-model workflows and cross-OS desktop automation. c/ua also allows easy sharing and distribution of VM images for collaboration. c/ua enables AI agents to control full operating systems in high-performance virtual containers with near-native speed on Apple Silicon. It supports agent loops such as UITARS-1.5, OpenAI, Anthropic, and OmniParser-v2.0. For developers, c/ua provides tools like Lume CLI for VM management, Python SDKs for agent development, and example code for direct control of macOS VMs.
Learn more
