# OpenAI API

Pensieve is a privacy-focused passive recording project that automatically records screen content, builds intelligent indexes, and provides a convenient web interface for retrieving historical data. Inspired by Rewind and Windows Recall, Pensieve differentiates itself by allowing users complete control over their data, avoiding transmission to untrusted data centers. Key advantages of Pensieve include easy installation, complete data control, support for full-text and vector searches, integration with Ollama, compatibility with any OpenAI API model, support for both Mac and Windows (with Linux support in development), and extensibility via plugins.

WebLLM is a high-performance in-browser language model inference engine that utilizes WebGPU for hardware acceleration, enabling powerful language model operations to be executed directly in web browsers without server-side processing. This project aims to seamlessly integrate large language models (LLMs) into the client side, resulting in cost reduction, enhanced personalization, and privacy protection. It supports various models, is compatible with the OpenAI API, is easy to integrate into projects, and supports real-time interaction and streaming, making it an ideal choice for building personalized AI assistants.

o1-engineer is a command line tool designed to help developers efficiently manage and interact with projects using OpenAI's API. It provides functions like code generation, file editing, and project planning to simplify the development workflow.

gptpdf is a tool that uses large visual language models (such as GPT-4o) to parse PDF files into Markdown format. It utilizes the PyMuPDF library to identify non-textual areas and leverages the OpenAI API for content parsing, enabling near-perfect rendering of layout, mathematical formulas, tables, images, and charts. With an average cost of $0.013 per page, it offers both efficiency and affordability.

Sumr is a Safari browser extension that leverages the OpenAI API to rapidly generate web page summaries, helping users efficiently access the core content of articles. It operates without switching applications or requiring a ChatGPT subscription, utilizing local storage functionality to save time and API usage costs. Sumr supports cross-platform usage, including iOS, iPadOS, and macOS, and users can customize API keys for cost-effective implementation.

The Zero & All Things Large Model Open Platform is a platform that provides access to high-quality Zero & All Things large models through API calls. The Yi series models are based on the cutting-edge research achievements and high-quality data training of Zero & All Things, and have achieved SOTA performance on multiple authoritative lists. The main products include yi-34b-chat-0205, yi-34b-chat-200k, and yi-vl-plus models. yi-34b-chat-0205 is an optimized chat model with an improved instruction compliance ability by nearly 30%, significantly reduced response latency, suitable for chat, Q&A, and dialogue scenarios. yi-34b-chat-200k supports up to 200K long contexts and can process content of approximately 200,000 to 300,000 Chinese characters, suitable for document understanding, data analysis, and cross-domain knowledge application. yi-vl-plus supports high-resolution image input and possesses capabilities of image Q&A, chart understanding, OCR, and is suitable for analyzing, recognizing, and understanding complex image content. The API advantages of this platform include fast inference speed and full compatibility with the OpenAI API. In terms of pricing, new registered users receive a 60 yuan trial amount, yi-34b-chat-0205 is priced at 2.5 yuan per million tokens, yi-34b-chat-200k is priced at 12 yuan per session, and yi-vl-plus is priced at 6 yuan per million tokens.