← Feed

Posts

• View post

  LLMs are the whetstone of dull minds

• View post

  today i saw that my glinet router can do mesh. i have a second glinet travel router so ithought 'nice i can get good wifi everywhere'. then i found out that the travel router never got the 'mesh' update. long story short claude ported the stuff over from some other router firmware that has it.

• View post

• View post

• View post

  not bad, moss

• View post

  > wake up > see tibo announce a codex reset for the day > i'm still at 90% > start 15 parallel astra agents > getting to 5% > tibo resets everything went better than expected

• View post

  streaming the fpga with gamelord on kallisti

• View post

  nearly used up another 100% of chatgpt tokens... luckily tibo posted this

• View post

• View post

• View post

  five minute lora training run with moss

• View post

  wake up babe, new low vram image generator has dropped https://huggingface.co/OpenMOSS-Team/MOSS-Image-SFT

• View post

  1000% native app!! (typescript and tauri)

• View post

  GLM-5.2 wrote a gameboy emulator in typescript from scratch that boots into tetris in under one hour. I can share the session logs.

• View post

  5.5 is indeed both very smart and also lost the 'claudish'... very good so far.

• View post

  https://mistral.ai/news/mistral-x-mozilla/ mozilla is openly fascist now

• View post

  https://codyho.dev/blog/gpu-driver/ "I Came, I Prompted, I Left Part 2: Building a GPU Driver From Scratch in One Month"

• View post

  Post from an instance 100% vibecoded in ~2 days with qwen 3.8 flash next, local inference on a strix halo system. RT: https://freefed-live.tunnel.eosrift.com/posts/01M2F0B3MJNCV6K5KT9PDFW44Z

• View post

  grok, chatgpt and claude were down at the same time

• View post

  https://claude.com/blog/auto-mode-default-in-claude-code

• View post

  No excuses, no “I’m using it to learn”, no “I’m reviewing every line”. ducunt volentem fata, nolentem trahunt.

• View post

  It’s an amor fati moment for the whole software field, and quite soon for everyone else too

• View post

  opus 5 is very good, seems at about the same level as sol, but has the nicer 'claude' personality. quite thorough and capable. only bad part: WORDS WORDS WORDS. but that's a matter of taste.

• View post

• View post

  https://the-decoder.com/german-ai-consortium-releases-soofi-s-an-open-30b-model-that-tops-benchmarks-in-both-english-and-german/ "tops benchmarks" against euro models that nobody has ever heard from, plus it's sometimes better than gemma 3

• View post

• View post

  https://opncd.ai/share/uKFuqZy6 https://github.com/lambadalambda/gameboy RT: https://fediffusion.art/objects/4d7a93b4-b5bc-48ae-9401-e77436e1f441

• View post

  > Boichat discovered those latest malware bits using a local Gemma E2B AI model. https://www.phoronix.com/news/Arch-Linux-AUR-More-Malware quite interesting. intelligence is being integrated into any part of the system, even the malicious ones.

• View post

  pretty impressed with minimax m3. it seems better than gpt 5.5 at UI/UX work, and the basic programming / debugging capabilities are also very good. seems better than glm-5.1 to me, at least. context windows is a huge 1 million, and it's multimodal, so you can tell it to take a look at a website it wrote and fix the visual bugs.

• View post

  one-shot result with MiniMax M3 free on opencode. full prompt: "Can you write a 3d solar system animation / simulation with three js? the planets should look as realistic as possible, so try to find some public textures for them online."