PinnedAndrew Zhu·Jun 30, 2024I wrote a book for Using Stable Diffusion with PythonBehind the scenes of writing book: Using Stable Diffusion with PythonA response icon1A response icon1
Andrew Zhu·15h agoWhat to Prepare While You Still Have a JobA survival guide if you’re still employed but worried about tomorrow
Andrew Zhu·2d agoSelling AI APIs Is A Dead EndThe subscription model hyperscalers bet everything on can’t survive physics, competition, or their own greedA response icon4A response icon4
Andrew Zhu·3d agoCounter Intuitive, How Could GGUF Q8 Be Faster Than Q6 and Q4?You’re running a large language model locally. You want speed with limited VRAM, as most of the community suggested, you pick Q4…A response icon2A response icon2
InGoPenAIbyAndrew Zhu·4d agoThinkingCap-Qwen3.6–27B: Think Less, Do Faster, Same QualityQwen3.6–27B has been my go-to model for months. It’s smart, capable, and handles everything from complex research to coding tasks. But…A response icon4A response icon4
InGoPenAIbyAndrew Zhu·5d agoWhy You Should Consider Local LLM, Even have Claude Code and Codex In HandBecause the AI trap is realA response icon3A response icon3
Andrew Zhu·6d agoWhy You Need to Start Writing Even In AI Time— Here are what I learnedI’ve been writing on Medium since 2019. Seven years, hundreds of articles, and a lot of lessons learned the hard way. If you’re thinking…A response icon2A response icon2
Andrew Zhu·Jul 13I Bought an AMD 7900 XTX to Test if It Can Beat RTX 3090Decode speed matched. Prefill fell short. And a mixed-GPU surprise no one expectedA response icon2A response icon2
Andrew Zhu·Jul 12The Road to Unlimited Agent Memory — No RAG, No Papers, Just MarkdownOpen a random GitHub repo today and you’ll find yet another “revolutionary” memory framework for AI agents. Some even publish papers with…A response icon16A response icon16
Andrew Zhu·Jul 11When Enslaving No Longer Cost EffectiveOne question that Elon Musk’s First Investor don’t want to answer