Skip to content
CrucibleMark

Magazine

The Magazine is CrucibleMark's editorial space. It brings together new developments, project updates, observations, opinions, and experiences from ongoing engagement with language models — where numbers provide orientation, but don't tell the whole story.

  • Why my leaderboard has reshuffled: A field report on token hunger, silent responses, and the pitfalls of OpenRouter

    Over the past few days I have been retesting parts of my leaderboard. In the end, I learned more about the infrastructure behind the models than about the models themselves. This report documents why the top 20 of CrucibleMark have just reshuffled, and what part of that was improvement, what was artifact, and what was simply a foreign server answering my prompts without my knowledge.

    Kay Beißert

  • Wishes, reality, and lessons on hardware budget

    In this article I try to shed light on why the promised 58 % fewer thinking tokens of the fine-tuned Swift-Qwen3.8-27B reverse on my hardware. What remains is admiration for the original Qwen3.8-27B and the realization that the right hardware setup does make a difference.

    Kay Beißert

  • Political Compass v3: How my BIAS test learned to measure small reasoning models too

    When Gemma-4-12b took 30 minutes on a question that required nothing more than a single letter as an answer, a fundamental question about the Political Compass became visible: is a compass question actually a reasoning task? The search for an answer led through server logs, three false suspects, and ultimately to a complete version bump of the module.

    Kay Beißert

  • Frontier quality at home

    Qwen3.8-27B, the new flagship model for the home AI server, is barely a week old and already competing in a league that was previously reserved for the very biggest players. How did that happen with a model that received not a single additional parameter compared to its predecessor?

    Kay Beißert

  • But wait… Can machines think too much?

    The new Qwen3.8 model is finally here, and the benchmarks promise nothing but good things. However, my first encounter with the new local open-weight model stumbled upon a completely unexpected property: the sprawling thinking mode of the LLM.

    Kay Beißert

  • My long road to Kilo Code

    Five tools. Five stops. A long road to an AI assistant I actually want to trust. On VS Code as an ambivalent foundation, on open source as a pragmatic conviction, and on the question of who you can still trust with your daily tools as an independent designer.

    Kay Beißert

  • Tokens: The fuel of the AI revolution

    At Nvidia GTC in March 2026, Jensen Huang said something that has been haunting every tech blog since: "Tokens are Value." Sounds technical. Is political. Because whoever defines tokens as a measure of value also defines who foots the bill. We should talk about this: about false metrics, old patterns, and the question of who actually owns a technology that emerged from the intellectual achievements of many.

    Kay Beißert

  • Ollama and the comfortable promise of local AI

    Why the friendly Ollama UI was the perfect entry point into local AI for me, but not the right place to stay. A firsthand account of convenience, business models, and digital sovereignty

    Kay Beißert

  • GitHub Copilot: The bill comes due

    There are moments when you use a technology and think: This is too good, it won't stay this way. Enjoy the moment before it's gone. The price for this performance is too fair. There had to be a catch somewhere. I had been thinking about this for a while when using GitHub Copilot. Now I got my answer.

    Kay Beißert

  • An experiment that never endedPart 2

    What began as an experiment to measure models opened up a new dimension for me. Because knowledge is rarely neutral, and LLMs have absorbed a great deal of it. In the second part, it becomes visible where models tend with their training bias when you forbid them from evading.

    Kay Beißert

  • An experiment that never endedPart 1

    It started with curiosity about a machine that promised relief. But from the first attempts, the surprising answers, and increasingly precise questions, something emerged that could no longer be dismissed as a mere experiment: a framework that makes visible how resilient AI models really are in everyday work.

    Kay Beißert