WebSailor is an open-source network agent from Alibaba's Tongyi Labs, focusing on complex information retrieval and reasoning tasks. It utilizes innovative data synthesis methods (such as SailorFog-QA) and training techniques (such as rejection sampling fine-tuning and DUP...).
Furion is a free and open-source development framework based on .NET that simplifies the development process and improves development efficiency. It supports multiple operating environments, including Windows, Linux, Mac OS, and Docker, and is suitable for developing Web APIs, Web applications, etc.
Hunyuan3D-PolyGen is the industry's first art-grade 3D large-scale model generator, launched by Tencent's Hunyuan team. Combining its self-developed high-compression-ratio representation technology (BPT), it can generate complex geometric models with tens of thousands of faces, supporting triangular faces...
Gen-CLI is a command-line AI programming tool based on the open-source Gemini-CLI and developed using the SiliconCloud platform API. Gen-CLI provides domestic developers with efficient programming capabilities similar to Gemini-CLI. ...
MetaStone-S1 is a reflective generative large-scale model launched by MetaStone Technology, which is the first to integrate deep reasoning with the self-selection capability of the reasoning chain. The core of the model uses a self-supervised reflective paradigm, based on a policy model and a process scoring model with a shared backbone...
HumanOmniV2 is an open-source multimodal reasoning model from Alibaba's Tongyi Lab. The model addresses multimodal reasoning challenges based on a forced context summarization mechanism, a large-model-driven multi-dimensional reward system, and a GRPO-based optimization training method...
Skywork-R1V 3.0 is an open-source multimodal reasoning model from Kunlun Wanwei, possessing powerful cross-modal reasoning capabilities and interdisciplinary generalization abilities. The model achieved a high score of 142 in the National College Entrance Examination (Gaokao) mathematics test and performed well in the Multidisciplinary Reasoning Evaluation Model (MMMU)...
Grok 4 is xAI's latest large-scale AI model, boasting a 10-fold improvement in reasoning ability compared to its predecessor. The model possesses exceptional reasoning capabilities, achieving near-perfect scores on challenging exams like the SAT and GRE, and performing well on multiple benchmark tests...
TradingAgents-CN is a Chinese financial trading decision-making framework based on a multi-agent large model. Developed on top of TauricResearch/TradingAgents, it provides Chinese users with a complete documentation system and local support...
PromptForge is an AI-powered tooltip engineering workbench that helps users efficiently create, test, and optimize AI-generated cues. The platform provides AI-assisted cue building and optimization suggestions based on intelligent generation capabilities, and features advanced analytics...
RoboBrain 2.0 is a powerful open-source embodied brain model that unifies perception, reasoning, and planning, supporting the execution of complex tasks. RoboBrain 2.0 includes two versions: 7B (lightweight) and 32B (full-scale), based on heterogeneous...
RoboOS 2.0 is an open-source, cross-ontology brain-brain collaborative framework developed by Zhipu, specifically designed for embodied intelligence. The framework supports multi-robot collaboration, achieving lightweight deployment based on the integrated MCP protocol and serverless architecture, thus lowering the development threshold.
EXAONE 4.0 is a self-developed hybrid reasoning model launched by LG AI Research in South Korea. The model integrates general natural language processing and advanced reasoning capabilities, and supports Korean, English, and Spanish. The model is available in a 32B professional version and a 1.2B version...
Voxtral is an advanced audio model from Mistral AI, leveraging its superior speech transcription and deep understanding capabilities to advance speech as a natural form of human-computer interaction. Voxtral is available in 24B and 3B versions, suitable for...
Excel MCP Server is a server tool based on the Model Context Protocol (MCP) that allows you to manipulate Excel files using an AI Agent without requiring Microsoft Excel to be installed. Excel MCP Server...
Seed GR-3 is a general-purpose robot model launched by ByteDance's Seed team, which has high generalization ability, long-range task processing ability and flexible object manipulation ability.
OxyGent is an open-source multi-agent collaboration framework developed by the Oxygen team at JD Retail. It helps developers efficiently assemble multi-agent systems. OxyGent abstracts tools, models, and agents into pluggable modules (Oxy), supporting features like...
Seed LiveInterpret 2.0 is an end-to-end simultaneous interpretation model developed by ByteDance's Seed team, supporting bidirectional Chinese-English translation. It boasts near-human translation accuracy and extremely low latency, enabling "speaking while listening"...
KAT-V1 is Kuaishou's open-source AutoThink model, available in versions 40B and 200B. The model integrates thinking and non-thinking capabilities, automatically switching thinking modes based on the difficulty of the question. Version 40B's performance is close to that of Deepin...
Qwen-MT is a machine translation model developed by the Alibaba Tongyi Qianwen team, based on the powerful Qwen3 architecture. The model supports high-quality translation between 92 languages, covering over 95% of the global population, and can meet diverse cross-language needs...
MonkeyCode is an open-source, enterprise-grade AI programming assistant that supports private deployment and offline use, ensuring code privacy and security. MonkeyCode is compatible with various large local models and features code completion, natural language programming, ...
SuperDesign is an open-source AI design agent that helps designers and developers generate UI prototypes, components, and wireframes directly within their IDEs. The tool supports natural language input and can generate multiple design options in parallel, allowing users to quickly...
Agentar-Fin-R1 is a large-scale language model launched by Ant Financial specifically for the financial sector, enhancing reasoning capabilities, credibility, and domain expertise in financial scenarios. The model is developed based on the Qwen3 base model and provides 8B and...
Step 3 is the latest generation of fundamental large-scale model released by StepStar, designed specifically for the inference era, combining high performance with extreme cost-effectiveness. Utilizing the MoE architecture, it boasts 321B total parameters and 38B activation parameters, making it the first...