Michael Christen

Open Source and Search Engine Developer, AI Research, Maintainer of Also on

[Michael Christen @orbiterlab]

FrOSCon 2026: Baue deine eigenen KI-Agenten -
Mitschnitt vom Workshop zu den KI-Agenten-Tools aus https://t.co/I5cxSsvivW
LLMs abfragen programmatisch auswerten, Linux-typische kognitive Filter bauen und 1-Script Coding Agents schreiben..
https://t.co/tJ7rEvbD5M

[Michael Christen @orbiterlab]

Email was once a resilient, decentralized infrastructure, now look at this MS cloud shit

[Michael Christen @orbiterlab]

gpt-6.1-sol-think saturates PE-Bench-200-Java from https://t.co/8iF4I8uYEv: Python 99%, JavaScript 99.5%, Java 100%, Rust 99%, Clojure 87%. The first model in this benchmark to solve all of Project Euler’s first 200 problems from https://t.co/IDLclJMfiw

[Michael Christen @orbiterlab]

vcoding useless? game 4 YaCy...

[Michael Christen @orbiterlab]

mflux-server #image #generation for #mac with support for qwen-image-2.1 using mflux 0.20.0 https://t.co/bSqYNuxi0I
Start the server with
./run.sh --model qwen-image-2.1
Example below uses image editing of the #YaCy logo

[Michael Christen @orbiterlab]

mflux-server 0.19.2 https://t.co/bSqYNuwKba #imagegeneration #mflux #server

[Jimmy Wales @jimmy_wales]

@davidsirota How about 4: there's a crazy moral panic going on. AI is a powerful and mostly wonderful disruptive technology and humanity will be better for it.

[Michael Christen @orbiterlab]

ox-alpha has a super-human factor of > 100 in python, javascript and rust in the project-euler-llm-benchmark https://t.co/8iF4I8uqOX second-best tested right behind deepseek-v4-flash-0731. It rejected to be tested in no-thinking mode, the only model since GPT-OSS to reject this.

[Michael Christen @orbiterlab]

@bijanbowen I tried to use ox alpha with reasoning=none and it rejected the request. The only model that did this before was gpt-oss!

[Michael Christen @orbiterlab]

Do you know SketchForge? This is a free/open Tinkercad clone that runs on your own computer. It has a MCP server inside, I attached it to codex and prompted "create a house with garden and a car in front". This is amazing! https://t.co/fSTQWeXQXC

[Michael Christen @orbiterlab]

Cyberdeck?? Cyberserver!!

[Michael Christen @orbiterlab]

@markaferguson @Claude @opencode @Alibaba_Qwen Can you share your opencode.json configuration with Qwen3.6 please? I constantly run into situation where opencode does something for 10 minutes and then has no result at all. I might have missed something in my models set-up.

[Michael Christen @orbiterlab]

This is crazy: Qwen3.6 27B @Alibaba_Qwen is leading 6-times larger MiniMax-M2.7 https://t.co/AresagYlqd

[Michael Christen @orbiterlab]

Qwen3.5 / Gemma4 Comparison
https://t.co/8iF4I8uqOX

[Smoking Wheels @smokingwheels]

YaCy Release 1.941 - The AI Release https://t.co/aSJCP1SCVl

[Michael Christen @orbiterlab]

KI Chat in YaCy - Vortrag von den Chemnitzer Linuxtagen https://t.co/OiEQCqRLi9

[Michael Christen @orbiterlab]

YaCy 1.941 is out: a big step toward transparent and private AI. New chat + RAG workflows, tool calling, MCP integration, YaCy Packs, better ranking/snippets #YaCy #Search #RAG #AI
Release Notes: https://t.co/MmDDXmqvgE

[Michael Christen @orbiterlab]

a kanban board that works off tickets without a team?

[Michael Christen @orbiterlab]

Make life better

[Michael Christen @orbiterlab]

qwen3.5 is leading the non-thinking project-euler leaderboard https://t.co/VWl8PUwpAI @Alibaba_Qwen
- qwen3.5:122b-a10b is best for its size
- qwen3.5:35b-a3b is best for its size
and also
- qwen3.5:27b, qwen3.5:9b, qwen3.5:4b: best for its size