SuperAPI AI Directory
Partner with us中文AI newsFavorites

All AI tools

35,708 results · Page 69 of 992

DeepAnalyze

AI agents

DeepAnalyze is the first agentic large language model for autonomous data science, developed by teams from Renmin University of China and Tsinghua University, and positioned as an "AI data analyst." It can automatically complete the entire data science workflow without human intervention, including data preparation, analysis, modeling, visualization, and report generation. It can also conduct open-ended deep research on structured data (databases, CSV, Excel), semi-structured data (JSON, XML, YAML), and unstructured data (TXT, Markdown), ultimately producing analyst-grade research reports. The project's model, code, training data, and demo are all open source, allowing local deployment or secondary development of your own data analysis assistant. The product offers multiple ways to use it, including WebUI, JupyterUI, a command-line interface, and an OpenAI-style API. It supports vLLM inference deployment and provides quantization recommendations for different VRAM conditions.

Read more

Design Arena

AI design

Design Arena is a global crowdsourced design benchmarking platform that aggregates the world's top AI models. Users can describe what they want to create, such as a website, game, image, video, presentation, or app, and instantly explore how each model brings the idea to life. The platform covers millions of users across more than 190 countries and is used to discover new models, compare capabilities, and find the intelligent model best suited to a creative task. Users can also participate in challenges, vote, and crown winners.

Read more

docAnalyzer.ai

AI office

docAnalyzer.ai is a source-grounded AI document workspace for professional practitioners, focused on turning document overload into attributable deliverables. It offers three specialized chat modes: Ask docAnalyzer helps plan the next step, Focus chat performs in-depth analysis of specified sources, and Co-work chat refines output in a structured canvas editor. Users can use 30 models from 11 providers within the same subscription and can switch mid-conversation without restarting the agent or losing context. The system can generate downloadable outputs such as PDF, XLSX, HTML, charts, diagrams, and packaged files, and these outputs can be referenced across multiple turns of conversation, converted in format, or combined into new works. The platform emphasizes multi-turn retrieval, reading, refinement, and review; every citation can be clicked to jump to the original page or section, and it provides context consistency controls. The official statement says content is never used to train models, workspaces are isolated by tenant, and outputs are limited to the session scope. It supports PDF, Word, PowerPoint, ODT/ODP, RTF, EPUB, Excel, Markdown, plain text, HTML, and in experimental mode CSV and JSON; it can also fetch web pages via URL, with automatic OCR for scanned pages.

Read more