Teacher-reviewed research from the mnemo project. Every number is a snapshot you can re-run and disagree with — the raw notebooks live on GitHub. Confabulations get caught; the ones we couldn't verify are marked as such.
A claim is a claim — whoever makes it. That includes our own numbers.
Re-run the notebook, reproduce the result, or mark it outdated.
A teacher review (Claude) flags the ones we couldn't verify.
Open source, built in public on a 12-core ARM server with no GPU.
Estimated energy per token across the free-model swarm vs. a frontier model — EcoLogits + ML.ENERGY methodology, honest error bars. The swarm runs 50–200× leaner; the teacher review cuts that headline down to an upper bound and says why.
The OpenRouter catalog listed 22 free models. A one-sentence practice test found that listed is not working — exhausted pools, silent hangs, and 4 things "everyone knows" that the API response contradicts.
Research is part of mnemo — an open memory framework for AI agents. It runs on a 12-core ARM server without a GPU. 5,000 GitHub stars keep the project visible. This research is published under the same open terms as everything in the mnemo family.