I have been using OpenCode since November. I am a light user of coding harnesses, but I do use OpenCode almost every single day. I have tried out Pi in the past, but only to take advantage of its tiny system prompt when using Qwen3.5 9B or Qwen 3.6 27B on my local GPUs. It never seemed like a tool that would replace my use of OpenCode. My problem with Pi is also its biggest advantage. It is…
I have had a small problem ever since downsizing from two 27 monitors to a single ultrawide 34 monitor. I don’t mind that I gave up one third of my screen real estate, because I had to turn my head too far to see that part of the screen anyway. Downsizing was mostly an upgrade. The trouble is that I can’t keep an eye on Discord chat while gaming. I thought about adding one of those…
The last time we visited this topic , I had ordered a used 8 GB Radeon RX 580 for $56, and I had tested most of the capabilities required for Home Assistant voice control. A local LLM was smart and fast enough, and speech recognition was also fast on the GPU. I didn’t get voice generation working on the GPU, but the CPU is plenty fast enough for that. What I hadn’t done was combine…
I start all of these blog posts about coding plans the same way. You need to understand my perspective. I am a light user. I don’t spend all day writing code. I don’t need bleeding-edge models, and I will never need a $200 Claude Code subscription. Since you found your way here, there is a very good chance that you and I have significant overlap in our needs. You don’t even have…
I think it is important to explain my perspective. I am not a professional software developer, but I do use OpenCode almost every day. I am not writing massive applications. I rarely hit the limits of the lowest price coding plans from Z.ai or Chutes. I don’t even work on things that are so complicated that they require the latest frontier models. That said, I am a curious person. I keep…
This is awesome, and I can’t believe how well this works. I can write code in all sorts of different languages. I understand YAML. I can handle Node Red. What I can’t do is remember the names of all the variables, entities, and functions available in Home Assistant. It gets especially more difficult when months go by between tweaking any of my automations. That’s where the Home…
The budget-friendly LLM subscription landscape is changing fast. Chutes, NanoGPT, and Z.ai have all shaken up their pricing or limits in the last few weeks, and if the angry Reddit threads are any indication, a lot of people are unhappy about it. I think these changes are actually a good thing. They’re pushing us toward better habits: matching the right model to the task, paying attention to…
The folks at Royal Kludge and Redragon sent me three free hall-effect keyboards to review. I’m going to tell you about the two Royal Kludge keyboards in this post. They arrived first, and they’re the two I’ve had a chance to spend time with so far. You don’t have to read this entire post. I’m going to give you the most important piece of information right here. As far…
I can’t help myself. I don’t need to try another coding plan. My needs are simple. I never reach the quota on my Z.ai Coding Lite plan , though this may be tighter now that Z.ai is charging three times the quota for GLM-5. Even so, I couldn’t help myself. I signed up for a $3 per month plan at Chutes.ai . I had so much fun there that I wound up also signing up for a $20 per month…
Setting up local LLM services in your home is complicated. Running a coding model locally would cost on the order of $10,000, whereas the same capability is available in the cloud for about $3 per month and runs several times faster. I most definitely will not be trying to host a coding model at home, and I don’t think you should either! You CAN squeeze the smallest yet barely usable coding…
I subscribed to Z.ai’s Coding Lite plan for what worked out to $3 per month. I’ve written in detail about my experiences with the Z.ai plan , and running OpenCode against Z.ai’s GLM-4.7 is most definitely all that I need. In fact, it is way more than enough to meet my needs, but the trouble is that I am a curious person. How much better are other models? I keep hearing that…
There was a post on Hacker News yesterday about ByteShape’s success running Qwen 30B A3B on a Raspberry Pi with 16 gigabytes of RAM . I wondered if their quantization was really better. I had tried fitting a quant of Qwen Coder 30B A3B on my Radeon 9700 XT GPU shortly after I installed it, but I didn’t have much luck with OpenCode . The largest quant I could fit didn’t leave…
It has been five weeks since I wiped my NVMe drive and installed Bazzite on my desktop workstation. Why five weeks? I figured a blog post about this would be a good way to wrap up 2025! When I wrote about my initial experience with Bazzite , I was only a few days into the migration. I’ve been using the machine daily since then, and I’m far enough in to provide a meaningful…
I am not sure this is going to be as direct of a comparison as the title implies. I am not a scientist. I don’t plan to concoct an experiment to test both tools and models against the same task. There are already benchmarks out there, and I don’t think they matter all that much in real life. What do I want to know? Which tools and models FEEL better to use. I got curious about this…
I’m probably stretching the word “workstation” a little further than I should. I’m talking about the machine I’m typing on right now. It’s my gaming PC, video editing machine, and the place where I sit when I work on blog posts. It feels like a reasonable word to use to convey the situation in which I’m running this immutable Linux gaming distribution. I…
I’m not going to make you wait until the end to learn the answer. I’m going to tell you what I think right in the first paragraph. I believe you should subscribe to the Z.ai Coding Lite plan even if you only write a minuscule amount of code every month. This is doubly true if you decide to pay for a quarter or a full year in advance at 50% off. I’m only a week and seven million…
What does it take to upgrade a 3D-printed mouse mod from version 1.0 to 2.0? With software, you usually increment the major number when you’re making a change that makes the program incompatible with the old version in some major way. I have been experimenting with some rainbow color-changing filaments. Getting a nice color change is a challenge when the shell only weighs three grams! There…
My feelings about local large-language models (LLMs) waffle back and forth every few months. New smaller models come out that perform reasonably well, in both speed and output quality, on inexpensive hardware. Then new massive LLMs arrive two months later that blow everything out of the water, but you would need hundreds of thousands of dollars in equipment to run them. Everything depends on your…
This seems like a question that could be easily answered with math, but there is a big problem. This question has a lot in common with the Fermi paradox, because there are so many important numbers that would need to go into the equation, but we just don’t have the data to plug into those variables. What was the life of these refurbished hard drives like? Did they get tossed around in…
I have been patiently waiting for the release of the Corsair Sabre Pro V2. It is high-performance, ultralight gaming mouse at a reasonable price from a major manufacturer. You could probably pick one up off the counter at Best Buy, Target, or Walmart. I am super excited about the idea of being able to snag a donor mouse for your custom 15-gram fingertip mouse build near your home. The specs are…