Search LingjingHub
Search news, tools, models, tutorials, workflows, and community content.
20 results
A safer local Holo 3.1 agent workflow with llama.cpp and OpenClaw
A structured workflow for serving Holo 3.1 locally with llama.cpp and connecting it to OpenClaw, including model selection and agent safety controls.
Build a local real-time AI voice assistant on Windows
A cautious deployment guide for a local speech-to-text, LLM, text-to-speech, and browser UI pipeline on Windows, with hardware and security notes.
Claude Code Learning Resources, Official Docs, and Agent Skills
A focused directory of official Claude Code documentation, Chinese learning paths, and Agent Skills references.
Claude Code on Windows: Installation and Setup for Beginners
A beginner-friendly Windows setup checklist for Node.js, Git, VS Code, Claude Code installation, version checks, and PowerShell policy issues.
Codex Dream Skin Guide: Reversible Themes for Codex Desktop
Install, switch, verify, and restore Codex Dream Skin on macOS and Windows while understanding its local CDP security boundary.
Codex Learning Materials and Workflow Guide
A separate entry point for Codex learning materials, workflow habits, and project collaboration guidance.
Gemini API Embeddings Tutorial: Document vs. Query Representations
Based on Google’s official Embeddings documentation, this draft explains how the Gemini API represents text as vectors and why retrieval workflows distinguish document embeddings from query embeddings.
Gemini Embedding 2: An Official-Docs-Based Capability Brief
Google’s official Gemini API documentation describes Gemini Embedding as a way to convert text into vector representations for semantic similarity, retrieval, classification, and clustering. This draft separates documented facts from editorial interpretation and relies only on the supplied source.
Google Gemini Embedding API Docs: Using Task Types to Guide Text Embeddings
Google’s official documentation explains the Gemini Embedding API as a way to represent text as numerical vectors, with task types used to tailor embeddings to specific use cases.
Hermes Desktop: model connections, cross-platform use, and safety boundaries
A practical overview of the Hermes Agent desktop client, its hosted and local model connections, cross-platform workflow, and agent security considerations.
Hugging Face Spaces: A Tool for Hosting and Sharing Machine Learning Demos and Apps
Hugging Face’s official documentation describes Spaces as a way to host and share machine learning demos and applications, making it useful for presenting interactive ML projects.
Kimi K3 capabilities: vision, coding, and browser-based interface generation
A concise review of public Kimi K3 examples covering visual understanding, code generation, Three.js projects, browser UI recreation, and the limits of benchmark and cost claims.
MiniCPM5-1B: 128K context and practical local deployment choices
A practical look at MiniCPM5-1B, including its 1B scale, long context, reasoning modes, GGUF options, deployment paths, and benchmark limits.
Model Context Protocol Specification Defines Concepts for Connecting AI Apps With Tools and Data Sources
The official Model Context Protocol specification defines protocol concepts for connecting AI applications with tools and data sources. This draft does not infer adoption, performance, or product claims beyond the specification.
Ollama Official Documentation Scope Note
According to the official Ollama documentation, Ollama documentation focuses on running models locally and interacting with the local service.
Ollama: Official Documentation for Running Models Locally and Interacting with a Local Service
Ollama’s official documentation describes running models locally and interacting with the local service. It can serve as a basic reference for readers exploring local model workflows.
OpenAI Platform Documentation Overview: Understanding Platform Concepts, API Usage, and Model Integration
The OpenAI Platform official documentation is a primary source for understanding platform concepts, API usage, and model integration. This draft is based on the official overview page and avoids unsupported claims about pricing, quotas, benchmarks, or release timing.
OpenAI Responses API Model Integration Note
Based only on the official OpenAI Responses API reference, this draft summarizes how model integration should follow the documented API behavior without adding unsupported claims about capabilities, pricing, or release timing.
OpenAI Responses API: A Unified Interface for Model Responses and Tool Use
OpenAI’s official documentation describes the Responses API as a unified interface for model responses and tool use.
A Human-Reviewed MCP Editorial Workflow Based on Official Sources
This LingjingHub workflows draft proposes an editorial process that collects official sources, records evidence, separates facts from interpretation, and requires human review before publication. The referenced source is used only to support protocol concepts.