A model for everyone, built from scratch.
Blackwell is Centralia’s model: the name you will know the work by. Centralia Cognition Labs is an independent lab training Blackwell from scratch, for people, not for big labs or experts only. Step by step, in public, with the numbers on the page. Follow the build to try Blackwell, follow the run, or open the investor and grant briefing.
Be first to try Blackwell and follow each training step as it lands.
Honest eval gates at each step: a larger Blackwell is only trained when the smaller one finishes and its evaluations say the system is ready. Results are reported as measured, including where they fall short.
Safety guidance is part of Blackwell’s training material itself, so expected behavior is learned during training, not patched on afterward.
Blackwell is built around eight principles covering truth, harm, fairness, and accountability, the standards its answers are measured against.
Blackwell is being taught to work with documents: read a file, summarize it, and work on its contents, with the same safety principles applied to what it reads and writes.
The build has been shared openly all along, as it happened. Every daily Blackwell reading now lives on the interactive timeline below, one metric entry per day instead of a stack of graphics. Follow the trail, or support the next step:
Blackwell finishes
Training complete. The 1B run finished at step 19,074 of 19,074, all 10 billion tokens in, final loss 1.3996. Final checkpoint saved at 08:37 AM PST. Tuning is next, then the launch.
Slide your mouse or finger along the bar, tap a point, or focus it and use the arrow keys. Each stop is one day on the record. The ringed point is the latest verified update: the completed run.
Every part of this was built by one solo founder: the tokenizer, the data, the training system, the safety work, and the public updates. That is why five months is the honest timeline, and it is exactly what funding changes. Funding buys H100 compute and help, so the 3B Blackwell step and everything after it moves at the speed of the work, not the speed of one person and one machine.
I am Michael Smith. I taught myself this work, and I have ten years of building behind me. I did not wait for a lab to hire me or a team to hand me the hard parts. I built the lab.
Nobody handed me an education in this either. I learned it myself over ten years, through books, Instagram, Twitter, and YouTube, lesson by lesson, usually after work. School came later: I am a working adult now in my third year of a BS in Business Information Systems, with a career across several levels of technology behind me. I have also done contract work evaluating and training frontier AI models, the kind of work where you learn exactly how the best models are judged, and where they fall short.
Centralia is mine, built solo, from scratch: the custom tokenizer, the 19.5 billion token data diet, the training harness I wrote by hand, and Blackwell, the 1 billion parameter model that is now 100.0 percent trained. Nobody handed me a stack. I am the stack.
I measure everything and I post the numbers as they are, good or bad. That is the whole pitch: I ship, I measure honestly, and I keep going. What comes next is simple. Finish Blackwell. Run the eval gates and publish what they say. Add the safety work and file handling. Then scale to the 3B Blackwell.
— MICHAEL SMITH, FOUNDER, CENTRALIA COGNITION LABS
Tokenizer · 19.5B token data diet · hand built training harness · Blackwell, 1B, training complete
Self taught over ten years through books, Instagram, Twitter, and YouTube · a working adult with a career across several levels of technology
BS in Business Information Systems, Grand Canyon University, third year, started September 2024 · Google Data Analytics certification, completed October 2025
Evaluating and training frontier AI models
Syndero and Code Sonar
Finish Blackwell · eval gates · safety and file handling · then the 3B Blackwell
The cycle ends where it warms: amber, follow the build. Early access to Blackwell, investor briefings, grant and funding conversations, and research partnerships all start here.
CENTRALIA BLACKWELL · CENTRALIACOGNITIONLABS.COM · INVESTOR AND GRANT CONTACT: MICHAEL@CENTRALIACOGNITIONLABS.COM
All Blackwell figures on this page come from one verified daily log, the same entries the interactive timeline above reads: the latest is November 12, 2026, step 19,074 of 19,074, 100 percent, 10B of 10B tokens, final loss 1.3996, finished Thursday, November 12, 2026. Earlier daily entries carry their own verified readings (the November 2 monitor, the November 5 halfway mark, the November 6, November 7, November 8, November 9, November 10, and November 11 readings), so the run stats, the progress bar, and the timeline can never disagree. Timeline dates are the verified milestones listed in the log; the conception point is deliberately undated, about five months before November 2026.