Synthetic training data: where it works and where it fails
Public human text runs out between 2026 and 2032. Generated data fixes that only where a checker exists, which is a much narrower set of tasks than the pitch suggests.
Read More
Learn
Evergreen explainers, guides and comparisons. The reference material that stays true after the news cycle moves on.
How RLHF works, and what it costs the model
OpenAI found a 1.3B tuned model was preferred over 175B GPT-3. The three stages, the alignment tax, and whose preferences get encoded.
Read More
Mixture-of-experts explained: why parameter counts stopped mattering
A trillion-parameter model can cost the same to run as a small one, because only a fraction of it fires per token. How sparsity works and what it costs.
Read More
Open-weight vs closed AI models: what each one really costs
Fixed hardware cost against per-token billing, plus deprecation risk, data residency and the licence traps hiding inside models called open.
Read More
Reasoning vs standard AI models: when the extra cost pays
Reasoning tokens bill as output. A model thinking for 2,000 tokens before a 200-token answer charges you eleven times what you see.
Read More
AI model pricing compared: same benchmark, five times the bill
Grok 4.6 and GPT-5.6 Sol Max tie at 61 on the AA index. One charges $6 per million output tokens, the other $30. The maths of that gap.
Read More
AI scaling laws: has bigger stopped meaning better?
The curves have not broken. The recipe changed twice, the spending moved to inference, and the data stock has a date on it.
Read More
Context windows explained: what a million tokens actually changes
Advertised context and usable context are not the same number. What a million-token window actually buys you, and what it does not.
Read More
What Is Retrieval-Augmented Generation (RAG) and When Do You Need It?
How retrieval-augmented generation works, the two places it breaks, and when long context or a plain SQL query would serve you better.
Read More
How to Write Better AI Prompts: 12 Techniques That Actually Work
Twelve prompting techniques that have an actual mechanism behind them. And how to tell when prompting has hit its ceiling.
Read More