#DeepSeek
318 related articles
DeepSeek UI: An Open-Source Browser Extension That Supercharges the DeepSeek Web Experience
DeepSeek UI is an open-source browser extension for Edge and Chrome that adds keyword highlighting, one-click quoting, Markdown/HTML export, and auto-collapse thinking to DeepSeek's web interface.
A DeepSeek Researcher's Confession: The AI He Trained Is About to Replace Him
DeepSeek V4.1 kernel engineer Liu Shengyu admits the AI he trained will surpass his own skills within a year—yet he keeps pushing forward. Here's why.
DeepSeek Harness One-Month Retrospective: Highlights, Risks, and the Road Ahead
DeepSeek Harness one-month retrospective: Plugin ecosystem, Codex Kernel, supply chain risks, talent strategy, and the cloud Agent path ahead.
Open-Source Project Tested: Integrating DeepSeek and Other AI Models into WPS via API
An open-source project uses DSL to integrate DeepSeek and other AI models into WPS, enabling in-document generation and revision. ~8,000 tokens per 2,000 characters. Requires a paid API key.
DeepSeek V4.1 Flash Deep Dive: How an Encoder Architecture Slashes Inference Costs to Pennies
DeepSeek V4.1 Flash uses a hybrid encoder architecture to cut inference costs to ~$0.27/task. Deep dive into its 40-layer design, KV cache optimization, benchmark results, plus OpenAI's Navier-Stokes breakthrough controversy, Anthropic's slowdown call, and Mistral's record funding.
Build an AI Agent with Memory from Scratch: Full-Stack with Python, Flask, and DeepSeek
Learn how to build an AI Agent with persistent memory using Python, Flask, and DeepSeek API. Covers selective storage, on-demand recall, and cross-session memory.
Complete Guide: Local Deployment of Personal Knowledge Base with DeepSeek + RAGFlow
Step-by-step guide to deploying DeepSeek locally with Ollama and building a personal knowledge base using RAGFlow on Docker. Covers RAG, Embedding, and full setup.
Running DeepSeek V4.1 Flash Locally on Mac Studio: How Far Can Local LLMs Really Go?
A 24-hour stress test of DeepSeek V4.1 Flash on Mac Studio M3 Ultra: 55B params, N-gram mechanisms, thinking mode trade-offs, and Deep SWE benchmark analysis.
Complete Guide to Running Your Own Local DeepSeek: Web Access, Knowledge Base & Privacy
Step-by-step guide to deploying a private DeepSeek locally using Ollama, Chatbox, and AnythingLLM — with web access, RAG knowledge base, and full privacy.
DeepSeek + 3 Tools: Generate a Professional PPT in 5 Minutes
A step-by-step guide to generating a PPT outline with DeepSeek, then using Tongyi, Kimi, or Coze to create a professional presentation in under 5 minutes — all free.
DeepSeek Speaks Japanese: Observations on Natural Multilingual Interaction
A Bilibili demo shows DeepSeek conversing naturally in Japanese — with proper honorifics, emotional roleplay, and Chinese–Japanese switching. Here's what it means.
DeepSeek Said It Was 'Sleepy' While Transcribing Music? Let's Talk About Anthropomorphism in LLM Chain-of-Thought
A Bilibili creator found DeepSeek saying it was 'sleepy' mid-task. We explain why LLMs mimic fatigue, how Chain of Thought amplifies anthropomorphism, and what to do about it.
DeepSeek Harness in Practice: Building a Low-Cost AI Coding Powerhouse
Learn how to transform DeepSeek's open-source harness using Claude Code, Bright Data scraping, and vision models to build an AI coding workflow costing just half a cent per task.
How DeepSeek Upended Silicon Valley With Rock-Bottom Prices: The Collapse of AI Pricing Power
How did DeepSeek use open-source models and ultra-low pricing to shake Silicon Valley's AI dominance? From the 'DeepSeek Monday' market crash to the 2026 V4 launch — a full breakdown of AI pricing power collapse.
The Truth About Running DeepSeek v4.1 Flash Locally: 23 Seconds Per Token
DeepSeek v4.1 Flash tops benchmarks with an MIT license, but local deployment means 510GB storage, 23 sec/token on Mac Mini, and a $18,796 cluster that takes 4 years to break even.
DeepSeek V4.1 Flash Tops AA's New Benchmark, Surpassing Astra and Sparking Controversy
DeepSeek V4.1 Flash tops Artificial Analysis's new private benchmark in Intelligence Index v4.3, surpassing Astra — but two index changes in three days spark community debate.
Confessions of a Chinese AI User: What Critical Context Anthropic's Accusations Are Missing
A Chinese AI user on Reddit responds point-by-point to Anthropic's accusations, adding technical and market context about reseller fraud, VPN clustering, and model distillation.
TutorialsClaude Code Model Configuration & Cost Comparison: A Practical Money-Saving Guide
Detailed guide on configuring DeepSeek V4 Pro, Sonnet, and other models in Claude Code with real cost comparisons, environment variable setup, proxy solutions, and money-saving strategies for developers.
Deep DivesThe "Worse is Better" Philosophy of Large Model Design: Why Simple and Brutal Beats Refined and Complex
Analyzing the "worse is better" philosophy in large model architecture: why DeepSeek V4 dropped N-gram, why Transformer dominates AI, and three iron laws of simple, efficient model design.
TutorialsOptimizing ASR Startup Scripts with DeepSeek: Intelligent Detection Over Countdown Timers
Learn how to use DeepSeek to rewrite ASR local AI project startup scripts from countdown mode to intelligent output-based detection, with full steps and troubleshooting tips.