The guides archive. Each piece takes a model family, a benchmark, or a method and treats it as a subject in its own right — with the same standard of evidence we apply to the registry itself. Grouped by what the guide is about, not by when it was written.
Comparative deep-dives on model families — what leads on which benchmark, and at what cost.
What benchmarks say about agent behaviour, and what actually happens when you run them.
ASR and TTS compared on the numbers that matter: WER, MOS, latency and price per hour.
OCR, invoice extraction and visual document retrieval — where vision LLMs are finally displacing specialist pipelines.
Domain-specific playbooks — regulation, manufacturing, tracking — written for engineers who have to ship.
How to read the field honestly — and why scale tends to win.
Longer-form pieces written for a specific audience — researcher, practitioner, buyer.