CATEGORY · 23 ARTICLES

Comparisons

Capability ladder from a zero-tool Claude session to a full Claude Code session, priced in tokens

Claude Code vs Claude Desktop: The Difference, Priced

We priced the gap between an agent and a chat client on one machine: 17,709 tokens of tools and project context, 63.4% of a default Claude Code prompt.

Sep 210 min read

Direct comparison between AGENTS.md open standard and CLAUDE.md Anthropic format

Comparisons

AGENTS.md vs CLAUDE.md: Which Config Standard to Choose

Choosing between Anthropic's CLAUDE.md and the open AGENTS.md standard. Scoping, memory injection, and the single-source bridging strategy.

Aug 2411 min read

A comparison matrix with each cell coloured by whether it is measured, vendor-stated, or unknown

Comparisons

AI Coding Tools Comparison: 130 Cells, 78 With a Source

Ten tools, thirteen attributes, and a provenance class on every cell. Forty per cent of the table is a gap, and no tool has a complete row.

Aug 2113 min read

A source census beside three green build gates and one misspelled Tailwind class that passes all of them

Comparisons

AI Tools for Web Developers: What No Check Catches

99% of this site's source is read by an automated check. A misspelled Tailwind class passes all of them and ships to the browser doing nothing.

Aug 2112 min read

Two store rankings of AI coding tools side by side, with unmaintained entries flagged

Comparisons

Best AI Coding Tools 2026: Ranked by What People Install

Two stores, two completely different rankings, and both of them put unmaintained software near the top. Here is the measured shelf and the three signals worth ranking on.

Aug 2013 min read

One measured coding session priced on three vendor cards as the session gets older

Comparisons

Best AI Model for Coding Compared: GPT, Claude, Gemini

Three vendors, one 354,835-token session, and three different pricing shapes. Gemini overtakes Claude Opus after 80 minutes, and the winning row rests on a rate nobody publishes.

Aug 1913 min read

Seven quantised coding models measured against one laptop's RAM and VRAM ceiling

Comparisons

Best Local LLM for Coding: Fit Decides It, Not Quality

Seven Q4_K_M candidates, real byte counts from Hugging Face, one ordinary Windows laptop. Zero fit the GPU, four fit RAM, and three separate runtimes refused to start.

Aug 1812 min read