Tag
#LLM
7 posts and news items use this tag.
Gemini 3.5 Live Translate Hands-On: New Two-Way Real-Time Voice Translation and Live API Developer Guide
Google has released the new Gemini 3.5 Live Translate model, supporting low-latency two-way real-time voice translation across more than 70 languages. This article walks through a hands-on test of the Google AI Studio web experience, explains how it works, and provides Live API development examples and WebSocket connection setup.
LLM Wiki Hands-on: Building a Personal Offline Knowledge Base That Large Models Can Organize and Update Automatically
Do your notes always feel tedious to organize? LLM Wiki is a personal knowledge base tool built around ideas from Andrej Karpathy. It can automatically read documents, run a two-stage chain-of-thought analysis, generate interconnected Wiki pages, and work well with Obsidian and local AI models.
Headroom Hands-On: A Context Compression Layer for AI Agents That Cuts Token Costs by 60–95%
Headroom is an open source context optimization tool built for AI Agents. It combines multiple compression algorithms, supports local reversible decompression (CCR), and can sharply reduce LLM compute costs while improving response speed.
CC Switch Extended Use: Testing Phone Remote Control for Your Computer and Codex Enhanced Mode
How can you use the exclusive features in CC Switch v3.16.1+ to keep Codex's official mobile login state and remote-control capability, while routing the actual model requests to third-party APIs such as DeepSeek? This hands-on guide walks you through an AI Agent setup where your phone controls your computer.
CC Switch Hands-On: A Local Routing Tool for Switching AI CLI Models with Your Own API Key in One Click
When using Claude Code or Codex in the terminal, how can you freely switch models, bring your own API key, or use DeepSeek? This hands-on article tests the CC Switch routing tool and walks through how to proxy AI CLIs while managing costs and MCP.
DeepSeek V4 Pro x Hermes Agent: Self-Written Intro Review and Deployment Guide
I asked Hermes to write this itself. 1M context, $0.87 per million tokens, cross-platform deployment, autonomous screenshots, writing, and git push. This article was handled by AI from research to publishing.
NVIDIA NIM Free DeepSeek-V4-Pro API Application and Multi-Framework Integration Guide
No credit card required, no trial period! This guide shows you how to connect to the newly released DeepSeek-V4-Pro and Flash models through NVIDIA NIM with minimal friction.