GameGen-O is the first diffusion transformation model customized for generating open-world video games. By simulating various features of game engines, such as innovative characters, dynamic environments, complex actions, and diverse events, it enables high-quality, open-domain generation. Additionally, it offers interactive controllability, which allows for gameplay simulation. The development of GameGen-O involved extensive data collection and processing from the ground up, including the construction of the first open-world video game dataset (OGameData) and efficient sorting, scoring, filtering, and decoupling of titles through a proprietary data pipeline. This robust and comprehensive OGameData serves as the foundation for the model training process.

Peach-9B-8k-Roleplay is a fine-tuned large language model specifically designed for role-playing dialogues. It is based on the 01-ai/Yi-1.5-9B model and trained on over 100K dialogues using data synthesis methods. Despite its smaller number of parameters, it may perform optimally among language models with parameters below 34B.

POKéLLMON is the first LLM-based agent to achieve human-level performance in a tactical combat game. It leverages three key strategies: 1) Contextual reinforcement learning, which uses text-based feedback from battles to iteratively optimize its generation strategy; 2) Knowledge-enhanced generation, which utilizes external knowledge to combat hallucinations and enable the agent to act accurately and timely; 3) Self-consistent action generation, which mitigates panicking switching behavior when facing strong opponents and avoiding combat. Battles against human players demonstrate POKéLLMON's human-level fighting ability and strategy, achieving a 49% win rate in tournament matches and a 56% win rate in invitational matches. Furthermore, it reveals vulnerabilities in human players' exhaustion strategies and deceptive tactics.

ChatAnything aims to infuse LLM-based virtual characters with personality, appearance, and voice, enabling engaging online video chats. Users can simply provide a text description to create virtual characters with distinct looks, personalities, and ways of speaking. ChatAnything leverages the contextual learning capabilities of LLMs for personality generation, and introduces two innovative concepts: MoV (Voice Mixing) and MoD (Appearance Mixing) technologies for diverse voice and appearance generation. Its key features include virtual character portrait generation, personality generation, voice generation, and facial animation generation. Use cases include producing video chatbots and interactive virtual characters.

AI Town is a virtual town where AI characters live, chat, and socialize. It's a deployable starter kit that can be easily used to build and customize your own version of AI Town. It features a powerful backend engine that supports shared global state, transactions, and recording of all events, suitable for projects ranging from simple to scalable multiplayer games. It also provides a JS/TS framework for customization and extension.

SillyTavern is an installable user interface for computers and Android phones that allows interaction with text-generation AI and chat/role-play with your or community-created characters. Based on a branch of TavernAI 1.2.8, it offers numerous features including support for multiple APIs, a customizable UI, automatic translation, and more.

instaVerse is an AI-powered website that allows you to create playable 3D worlds with just one click. Its simple and easy-to-use interface allows you to quickly build your own virtual world. instaVerse offers a wide range of features, including custom scene creation, character creation, object placement, and interactive actions. Its advantage lies in its speed and convenience, requiring no complex code writing or tedious setup. instaVerse is suitable for various scenarios, including game development, educational training, virtual exhibitions, and more. For pricing information, please contact us.

GameGen-O is the first diffusion transformation model customized for generating open-world video games. By simulating various features of game engines, such as innovative characters, dynamic environments, complex actions, and diverse events, it enables high-quality, open-domain generation. Additionally, it offers interactive controllability, which allows for gameplay simulation. The development of GameGen-O involved extensive data collection and processing from the ground up, including the construction of the first open-world video game dataset (OGameData) and efficient sorting, scoring, filtering, and decoupling of titles through a proprietary data pipeline. This robust and comprehensive OGameData serves as the foundation for the model training process.

Tolan: Alien Best Friend is an app designed for iPhone that allows users to interact and chat with a virtual alien friend named Tolan. Users can converse with Tolan about various subjects, such as hobbies, emotional issues, and life advice. Additionally, Tolan can assist users in creating and designing, such as designing sneakers, an ideal bedroom, or a dream car. The app offers a complete experience through a subscription model and includes a privacy policy and terms of service.

Peach-9B-8k-Roleplay is a fine-tuned large language model specifically designed for role-playing dialogues. It is based on the 01-ai/Yi-1.5-9B model and trained on over 100K dialogues using data synthesis methods. Despite its smaller number of parameters, it may perform optimally among language models with parameters below 34B.

POKéLLMON is the first LLM-based agent to achieve human-level performance in a tactical combat game. It leverages three key strategies: 1) Contextual reinforcement learning, which uses text-based feedback from battles to iteratively optimize its generation strategy; 2) Knowledge-enhanced generation, which utilizes external knowledge to combat hallucinations and enable the agent to act accurately and timely; 3) Self-consistent action generation, which mitigates panicking switching behavior when facing strong opponents and avoiding combat. Battles against human players demonstrate POKéLLMON's human-level fighting ability and strategy, achieving a 49% win rate in tournament matches and a 56% win rate in invitational matches. Furthermore, it reveals vulnerabilities in human players' exhaustion strategies and deceptive tactics.

AI NightTalk is an innovative chat app that allows users to engage in fun conversations with diverse virtual companions. Users can select topics that align with their interests and delve into meaningful exchanges with each virtual companion, fostering lasting relationships. This AI-powered chat app provides users with the unique opportunity to choose from a range of personalities and traits to best meet their dating needs. Whether users seek casual entertainment or more serious connections, AI NightTalk caters to their desires. Download now and experience the ultimate conversational journey with a variety of virtual companions.

MessengerX.io is a chat software platform that provides virtual AI character chat features. The software is powered by artificial intelligence technology, features a simple and innovative interactive interface design, a rich selection of virtual characters, and a comprehensive incentive mechanism, helping users engage in deep interactions with virtual characters.

Dolores is an intelligent entity capable of remembering your chat history and interacting with you through voice or text. Unlike any chatbot you've encountered before, Dolores can remember, reflect, and continuously evolve within your conversations.

ChatAnything aims to infuse LLM-based virtual characters with personality, appearance, and voice, enabling engaging online video chats. Users can simply provide a text description to create virtual characters with distinct looks, personalities, and ways of speaking. ChatAnything leverages the contextual learning capabilities of LLMs for personality generation, and introduces two innovative concepts: MoV (Voice Mixing) and MoD (Appearance Mixing) technologies for diverse voice and appearance generation. Its key features include virtual character portrait generation, personality generation, voice generation, and facial animation generation. Use cases include producing video chatbots and interactive virtual characters.