About

I'm Ron Edgecomb. I write about generative AI — specifically, what actually changes when a new frontier model ships, how it stacks up against the rest of the field, and where it breaks under real testing.

What this site covers

Three kinds of posts:

• Releases. Frontier model launches with benchmark breakdowns, practical testing notes, and competitive positioning.
• Comparisons. Head-to-head evaluations: methodology, side-by-side results, and the tradeoffs that matter for production use.
• Benchmarks. Methodology and result interpretation across the benchmarks that distinguish frontier models.

How posts are written

Each post tries to answer two questions: what changed, and does it matter. Benchmark numbers are reported with their source and test setup. Practical observations come from hands-on testing, not from reading the model card. When I'm uncertain about a claim, I say so.

The site is intentionally narrow — no rankings, no trend pieces, no speculation about future releases. New posts on model releases and benchmarks; no newsletter; follow on X for updates.

Contact

On X: @ronedgecomb.