A new OneLLM Worker card on the chat home screen. Cleaner navigation and improved local streaming replies.

Meet Pumli 3.0 — a more personal space for local AI. • A redesigned assistant with conversation history and editable preferences. • On-device model choices, including MiniCPM-V 4.7 35B and compatible models. • Multi-step Research with progress, sources and adjustable depth. • Optional connectors with clearer setup and tool approval controls. • Editable email drafts before opening Mail. • Online image creation and Seedance video tools using your own provider account or key. • Refined layouts for larger displays and improvements to everyday interactions.

1. Updated online models, including Gemini 3.5, GPT-5.5, and more. 2. Improved the overall user experience.

1. Added support for the **Qwen3.5 series local models**. 2. Improved the **connection experience for the OpenClaw module**.

1. Added the new Qwen Online Chat module. You can now access the most powerful model in the Qwen 3.5 series. 2. Introducing one-click deployment and configuration for the OpenClaw module, making your connections simpler than ever. 3. Added Dark/Light mode support for selected modules. 4. Fixed known issues and improved performance. Starting with version 2.0.3, OneLLM Pro is evolving into OneAGI — now with a bold new name, a fresh logo, and a much bigger vision. This marks the beginning of an exciting new chapter. Stay tuned — we’ll be unveiling what’s next very soon.

Updated the OpenClaw node — you can now control the device’s browser and other functions through conversation. Added Day/Night mode switching to multiple modules. Added Z.ai and Minimax to the Multi-LLMs module. Fixed known issues.

1. Introducing OpenClaw Node support – now you can pair OpenClaw with OneLLM Pro for a more flexible intelligent Agent experience. 2. Added online APIs for Z.ai and MiniMax, plus updates to multiple models. 3. The version 2.0 features a fresh interface and optimized user experience.

1. Added the latest online and local models including ChatGPT 5.2, Grok 4.1, and more 2. Fixed some known issues

1. Updated to the latest models for the Online API 2. Fixed some known issues

1. Added Google's latest Nano Banana (Gemini 2.5 Flash Image) model. 2. Improved and updated the AI voice chat module with OpenAI‘s Real Time model. 3. Optimized and upgraded several other modules for better performance.

1. Added local LLMs, including Gemma 3 270M. 2. Updated AI Video Module with new video generation, replacing the old one. 3. Fixed some known issues.

1. Added Google search capabilities to the local model, enabling internet-based Q&A even with a 0.6B model. 2. Introduced support for connecting local small LLMs with practical tools via custom Model Control Protocol (MCP) to enhance functionality. 3. Added support for recent local models, including gpt-oss. 4. Introduced GPT-5 API for online models.

1. Added support for local multimodal models, now compatible with multiple local vision model chats. 2. Introduced several new online LLMs, such as Grok4, Gemini 2.5 Lite, and more. 3. Revamped the AI role module, now offering improved custom role options. 4. Fixed several known issues.

1. Optimized user experience.

1. OpenRouter module now includes PDF and Image analysis, as well as Web Search capabilities. 2. We’ve added AI Gym Assistant module, allowing you to learn about your fitness equipment and interact with an AI fitness coach through visual and language-based chats. 3. AI Smart Center search now features Grok’s Live Search function, providing more search options. 4. We’ve fixed several known issues and improved the user experience.

OneLLM v1.5.3 is about to drop, and here’s what’s new in this update: 1. Added an AI-powered marine life recognition module—take your OneLLM Pro to the aquarium or go diving! 2. Improved the user experience for Multi LLMs and fixed some known bugs. 3. Enhanced the experience for Local LLMs, Dev Mode, and Online LLMs modules. 4. Integrated the latest models, including Gemini 2.5 Pro 05-20 and DeepSeek-R1-0528-Qwen3-8B-GGUF. 5. Fixed additional known issues and further streamlined the installation package size.

1. Fixed the issue.

1. Added AI Smart Center module, where you can access nearly all app features in a single interface, with support for LLMs to search the web and engage in conversations. 2. Introduced AI Voice Chat, allowing you to select different AI voices for continuous or one-off dialogues. 3. Continuously improved app functionality and fixed some known issues.

1. Added AI Food Calorie Module to better track your daily calorie intake and nutritional content. 2. Improved the AI PDF module, now with more model options to choose from. 3. Enhanced the UI and other details, fixing known issues.

1. Local LLMs now include the Qwen3 series models.

1. We've added new online LLMs like o4-mini and GPT 4.1, along with the latest GPT Image generation model . 2. We’ve added a Favorites feature—on the homepage, you can now add your AI tools to your Favorites. 3. We’ve enhanced the Multi LLMs feature for better context support and added press-to-copy functionality. 4. We’ve streamlined the app size for a smoother download and update experience. 5. We’ve fixed several known issues.

1. Multi-LLM Simultaneous Response Mode: OneLLM Pro v1.4.3 adds a new mode that lets you get answers from the world’s top LLMs with one click, saving you from constantly switching apps. 2. Study Assistant Feature: Introduces the AI Study Assistant, allowing answers from models via typing, snapping a photo, or uploading an image. 3. Upgraded Plant & Birds Module: The Plant & Birds identification module now supports saving custom prompts and more enhancements. 4. New Online LLMs: Adds Grok 3, Grok 3 Mini, and their Fast versions to the Online LLMs Module.

1. Added AI Plant & Bird Identification module. With spring in full bloom, you can now take OneLLM Pro along for a nature outing and discover a whole new side of spring! 2. Greatly improved the speed of local speech recognition and conversion, making AI transcription faster than ever. 3. Fixed some known issues.

1. AI image generation now includes the latest Grok and Gemini Image models, unlocking even more possibilities for your AI-crafted masterpieces. 2. Online LLMs get a boost with the addition of Gemini 2.5 Pro and Gemma 3 27b. 3. New icon shortcut feature—press and hold to jump straight into core features . 4. Fixed known issues.

1. Fully upgraded architecture and core engine, making the app more powerful and stable.