robin_rx

@oliverhenry Think its mostly hype and marketing. Many use OpenCode or Pi. I have tried Hermes too. I have yet to see an advantage in it.

I have still yet to meet someone actually using hermes. Whenever I mention it, i just get bots in my replies. Openclaw 4 lyf

@trq212 This is actually quite smart.

HTML is the new markdown. I've stopped writing markdown files for almost everything and switched to using Claude Code to generate HTML for me. This is why.

@badlogicgames Don't forget the Netzteil.

why? why shouldn't i be an absolute moron and buy this?

@badlogicgames The "neurosymbolic" of the poor man.

TIL about "caveman mode" to "save tokens". how many tokens in a session are actually model output? i think i'll become a gardener.

@brilliantlabsAR Was reading your website and still have no clue what your product can do. Great!

Look up. Look around. Play. Halo turns the world into a place worth exploring again. A new way to see, make, and play. brilliant.xyz.

Aaaaaaaaand GPT-5.5 is lobotomized. Nice! I seem to have some mental illness that let's me fall for this shit again and again. Anyone else experiencing planting pot level experience with this miserable clanker?

@0xSero @kilocode I think you have your answer right there.

@kilocode BYOK? I used Kilo when it was a VSCode plugin, I know you forked OC a while back. What would you say is your defining feature.

I still stand by Droid being the best agent harness out there. I’ve tested everything under the sun. #1 Droid #2 Pi #3 Amp #4 OpenCode #5 Codex CLI I am still working on a few reviews but performance wise this has been my experience.

@brendonovich @ErikBjare @ParthJadhav8 @nexxeln Interesting article. What was your reason to change to node.js?

@ErikBjare @ParthJadhav8 @nexxeln x.com/brendonovich/s…

OpenCode Desktop is now running on Electron. Let us know if you run into any issues :)

@0xSero What advantage do you see in Droid against OpenCode, Pi or Codex?

Warp - High level harness harness Droid - Main harness Pi - Local model harness Deepseek-V4-Pro - Main dawg Deepseek-V4-Flash - Local dawg GPT-5.5 - Backup dawg

@leftcurvedev_ That is brutal, man! Oh, I love this!

local ai haters in disbelief right now

Laser hits plasma field, plasma field shields matter from laser and focuses the laser into a high energy beam. This technology may be a ground breaking step to fusion energy.

Era of local coding agents has come.

@davepl1968 Depends on how the neural net is structured internally. I mean, if we create a complex model from scratch, there are infinite ways in how we can design its flow. So with one model you will get the same thing every time, while another one might provide broader responses.

A lot of people don't seem to understand that if you ask an AI the SAME question, you will get the SAME answer. Every. Single. Time. To avoid that, the models "salt" your prompt by adding random input params so that the model does not take the same path. But LLMs are not random or unpredictable - if you ask it the name of TinTin's dog, and it gives you an answer, then the same model with the same weights and the same input will ALWAYS say Snowy.

@1mustangcharlie @davepl1968 BMW X1 has one, though.

@davepl1968 Cars haven't used valve bodies like this in 20 years. It's all electronic solenoids and actual computer controls now. fluid pressure is still the control mechanism, but these clever "hydraulic analog computers" are long gone.

This a computer, and you likely own one. It's a hydraulic analog computer. It’s essentially a machined analog computer that computes with fluid instead of electronics: pump pressure is routed through passages that act like wires, while spool valves, springs, orifices, and check balls perform the equivalents of comparators, logic gates, delays, and one-way elements. What it “calculates” is the machine’s current operating state, whether conditions have crossed a threshold to justify changing state, how strongly to apply each output, and how quickly to make that transition without instability or shock. It does this by continuously balancing forces—pressure on different valve areas against spring preload and feedback pressure—so each valve shifts only when one hydraulic condition outweighs another, while restrictions and chambers add timing and smoothing. In plain English, it is a real-time fluidic state machine that solves “if this pressure is greater than that one, route flow here; otherwise hold, delay, soften, or override” entirely through geometry and oil. They're used in every car with an automatic transmission, where it makes choices like what gear to be in and how hard to apply clutches, etc.... And some dude worked it all out on paper back in the 1960s.

@Villm_3 @ollama @aiandcloud Why would you do that and not use llama.cpp directly?

@ollama @aiandcloud ollama很好啊 哪里唾弃了？？ 服务器部署ollama也有用呀

Ollama 是怎么样从开发者最喜欢的本地工具，一下子变得众人唾弃的呢 🤣

@KenamReemas @opencode Yes, I am gone again. Should have carefully read that a request is defined as about 150 tokens... meh. So one coding agent burst, kills a quarter of my usage quota. 😅

@rbn_rx @opencode I use it with open router. Its a token hog. Spent $5+ in ~4 prompts.

Qwen3.6 Plus and 3.5 Plus now available in Go - both strong - 3.5 is 3x cheaper - both support images - zero data retention update to the latest to try

@willreil The coolest thing about this clip is your euphoria and happiness about the success. I can relate to this feeling so much and as a software dev, I know that hardware is quite a challenge. But the dopamine kick is preem, when it finally works.

LETS GOOOOO IT WORKS!! This is my first functional hardware project, it solves a paint point I deal with daily. It feels so cool to have this own it’s own PCB that I manufactured. I am really proud of this project despite how simple it appears. Next up is making a case!

Oh, hello kids! Now it gets super interesting. @PrismML released new 1.58 bit models, and if they keep up the quality of the 1 bit Bonsai models, Prism will lead these ranks soon. (See link below).