The dependency list
Sources
Every outside document this site leans on, what it is cited for, which pages fall over if it changes, and when it was last actually read. This is the maintenance list, published rather than kept privately.
Published 12 August 2026 · Last verified 12 August 2026 · Every URL below was requested on that date and returned a working page
Why this page exists
On 12 August 2026 this site was found citing a real figure to two pages that did not contain it. The number was right and the citation was unfollowable, which is worse than no citation at all: a reader doing the check this site recommends would have concluded the figure was invented. That is logged in full, and this page is the structural fix. When every external claim is listed in one place with its source and its date, a citation that leads nowhere has somewhere to be caught.
It also makes the maintenance honest. The re-checking commitment on the changes page is easy to state and hard to verify from outside. What it actually amounts to is re-reading the documents below and seeing which pages move. There are sixteen of them, eleven of which carry figures. That is a real afternoon, not an impossible one, and now you can hold the site to it rather than take its word.
Primary sources: vendor documentation
These carry the hard figures. If one of these changes, something on this site is wrong until it is updated.
| Document | Cited for | Pages that depend on it | Last read |
|---|---|---|---|
| Anthropic — models overview | Context windows, output limits, prices, both cutoff dates, and the word/character approximations | Model facts, Tokens, Context windows, Training cutoff, Hallucination | 12 Aug 2026 |
| Anthropic — pricing | The Sonnet 5 rate becoming permanent; batch and caching discounts | Model facts | 11 Aug 2026 |
| Anthropic — token counting | The ~30% tokeniser increase and the model it starts at | Tokens | 12 Aug 2026 |
| Anthropic — token counting guidance for developers | The 15–20% tiktoken undercounting range |
Tokens | 12 Aug 2026 |
| Anthropic — prompting best practices | Examples as a steering lever, the colleague test, self-verification and its exception | Getting better results | 12 Aug 2026 |
| OpenAI — pricing | GPT-5.6 prices, and the long-context tiers that are not in the table | Model facts | 12 Aug 2026 |
| OpenAI — models | Context and output limits, knowledge cutoff | Model facts | 11 Aug 2026 |
| Google — Gemini API pricing | Gemini 3.6 Flash prices and tiering | Model facts | 11 Aug 2026 |
| Google — Gemini 3.6 Flash | Context and output limits, and the absence of a published cutoff | Model facts | 11 Aug 2026 |
| Meta — Model API pricing | Muse Spark pricing, the contributor tier, and the absence of a first-party Llama price | Model facts | 11 Aug 2026 |
| OpenAI — model optimization | What fine-tuning is sold as doing, in the vendor's own words | Fine-tuning | 11 Aug 2026 |
Vendor writing on mechanism
Not figures, but explanations of how something works, cited where this site describes the same mechanism. These age more slowly and are worth reading directly.
| Document | Cited for | Pages that depend on it | Last read |
|---|---|---|---|
| Anthropic — Building effective agents | The workflow/agent distinction, and the case for the simplest thing that works | Agents | 11 Aug 2026 |
| Anthropic — Reasoning models don't always say what they think | Visible reasoning not being a reliable account of the actual process | Thinking and reasoning, What AI is actually bad at | 11 Aug 2026 |
| Anthropic — Contextual retrieval | Why naive chunking loses the context a chunk needed | RAG and retrieval | 12 Aug 2026 |
| Anthropic — extended thinking | How a thinking budget is configured and billed | Thinking and reasoning | 11 Aug 2026 |
| Anthropic — prompt caching | Prefix matching, breakpoints, and what silently invalidates a cache | Prompt caching | 11 Aug 2026 |
One of these is aging, and it is worth saying so
The extended thinking document describes the mode that is being superseded. On Anthropic's own current model table, extended thinking reads "No" for its three newest models and "Yes" only for the oldest one listed, with adaptive thinking the other way round. Thinking and reasoning already describes the newer pattern correctly in its prose, so nothing on that page is wrong. But it is cited to the document for the older mode, and that is a citation that should move rather than a fact that should change. It has not moved yet, because moving it means reading the replacement properly first.
Learning resources
Listed on Start here as places to go, not cited for any claim. Each was opened and confirmed on the date shown; where a site does not publish pricing, that is recorded as a fact rather than guessed at.
-
Elements of AI
Free, non-technical, predates the chatbot boom. Confirmed 11 Aug 2026.
-
Anthropic Academy
Structured courses; publishes no pricing on its hub page. Confirmed 11 Aug 2026.
-
Hugging Face Learn
Hands-on, between conceptual and technical. Confirmed 11 Aug 2026.
-
DeepLearning.AI
Mixed free and paid; verify per course. Confirmed 11 Aug 2026.
-
Anthropic — recommended path for new developers
The primary source, already sequenced. Confirmed 11 Aug 2026.
-
fast.ai
For understanding the machinery rather than calling it. Confirmed 11 Aug 2026.
How to use this against the site
Pick any figure on Model facts, find the document it came from above, and open it. If the two disagree, this site is wrong and that is worth reporting. The dates in the last column are the honest limit of what is claimed: a source last read on 11 August is a source that could have changed on the 12th, and nothing here pretends otherwise.
The pattern to watch for is the one that caught this site out. A source can keep saying what it always said while quietly ceasing to be the right source — the extended thinking document above has not changed a word and is becoming the wrong citation anyway. Nothing about that looks like rot, which is exactly why it needs a list.
What isn't here
Research papers. Two pages on this site describe findings that are established in the literature without citing a specific paper, and both say so on the page rather than hiding it: Context windows on attention degrading over long contexts, and Hallucination on the mechanism itself. Those are real gaps. Filling them properly means reading the papers rather than citing whatever comes up first, so they stay flagged until that happens.
What changed, and when → Changes · The method → Is what you're reading out of date?