Analysis
The list price is not the price your agent pays
The model 5.7% cheaper on the price tag billed an agent session 4.4x higher. The cached-read rate decides an agent bill, and it runs from 2% to 25% of input.
Journal Independent editorial
The journal.
18 posts across analysis, opinion, how-to, research, and roundups.
18
6
16min avg
About
Independent analysis, opinion, and how-to coverage of LLM tooling -- written for engineers who pay the bill. No paid placements, no vendor talking points, no benchmarks lifted from press kits. Every model price comes from the vendor's own page, verified the week of publication; every cost calculation shows its math. The blog sits next to the price desk and the comparisons -- prose covers the questions a table cannot answer (when does a 90% cache discount stop mattering? which workload class breaks GPT vs Claude pricing parity?). Cornerstone analysis ships first; opinion, tools how-to, research summaries, and roundups will join as the catalogue grows. Read for the math, stay for the disagreements.
§ Filter by category
Category
Editor's pick
1 of 18 Showing 11 of 18 posts — page 1 of 2
Analysis 02
Running many agents at once: the tax is linear, the rebate is quadratic
Every extra agent pays for its own cold prefix. But re-reading history grows with the square of turns, so splitting one job across three agents ran 17% cheaper.
Tools Howto 03
The 20x in Claude Code caching is real — it's a penalty, not a discount
A cache miss costs exactly 20x the warm read of the same history. What that does to your plan limits, which settings change it, and the number nobody can compute.
Tools Howto 04
The Batch API discount isn't 50% -- and it isn't always a discount
Batch halves token rates at OpenAI, Anthropic and Google -- but takes 20% at xAI and nothing on its flagship. When batch beats a warm cache, and when it costs more.
Tools Howto 05
How to actually get prompt cache hits (and why yours are missing)
Prompt caching fails silently -- no error, no warning, just a full-price bill. The floors, the breakpoint rules, and everything that quietly invalidates a cache.
Analysis 06
Routing by task complexity: what the 2026 evidence actually shows
Ten routers launched on HN in five weeks. The peer-reviewed 2026 benchmarks say they barely beat a baseline. The arithmetic that decides it, on verified rates.
Tools Howto 07
The context engineering playbook: which lever, in what order, and what each is worth
Five levers, one deployment order: which context feature to reach for by symptom, what each does to your cache and your bill, and one config that runs them all.
Analysis 08
Token savings are not dollar savings: the conversion math
Cutting 63% of an agent's tokens moved its bill 9%. Cutting 1.2% moved it 15%. The arithmetic that converts token reductions into dollars, on verified 2026 rates.
Tools Howto 09
The MCP context tax: how to stop paying for tools your agent never calls
Five MCP servers burn ~55k tokens before you ask anything. Tool search and programmatic tool calling (both now GA) cut that 85%+ — with one caveat that bites.
Tools Howto 10
Compaction: when to summarize an agent's history — and when it backfires
Anthropic's compaction API summarizes an agent's history when it hits a token threshold. How it works, the billing pass you don't see, and when it backfires.
Tools Howto 11
Context editing: cut agent token bills ~84% by clearing stale tool results
Anthropic's context editing clears stale tool results from an agent's window, cutting token use up to 84%. How it works, the config, and the prompt-cache catch.
Roundup 12
Free AI compute in 2026: every free tier, trial, startup grant, and student deal — verified
Every legit way to get LLM compute for $0 in 2026: free API tiers, trial credits, student offers, and startup grants up to $350k — verified, with the catch on each.