ModelsClaude Fable 5 Guide: The First Mythos-Class Model, Still Served
Fable 5 launched in June 2026, was suspended within days, and now sits behind Fable 5.1. What it is, how its safeguards behave, and when staying on it makes sense.
ReadNotes on models, agents, and what inference costs
Practical writing for developers building with large language models — how they work, how to pick one, and how to keep the bill predictable.
Models
ModelsClaude Fable 5.1: What Changed From Fable 5 and Who Should Pay for It
Fable 5.1 kept the $10/$50 price, cut cache reads by 75% and added three breaking API changes. What improved, what breaks, and when Opus 5.5 is the better start.
Read
ModelsClaude Haiku 4.5 in Late 2026: When the Old Small Model Still Fits
Haiku 4.5 now costs more per token than its successor, Haiku 5.5. What it was built for, what breaks when you migrate, and the few cases where staying put is right.
ReadModelsClaude Opus 4.1 After Retirement: Where It Still Runs, How to Leave
Anthropic retired Claude Opus 4.1 on its own API in August 2026, but partner clouds still serve it. What it was, what it costs, and a checklist for moving off it.
Read
ModelsClaude Opus 4.5: The Release That Cut Opus Pricing by Two-Thirds
Opus 4.5 dropped Opus pricing from $15/$75 to $5/$25 and introduced the effort parameter. What it still does well, its 200K limits, and the retirement date ahead.
ReadModelsClaude Opus 4.6 Guide: 1M Context on the Last Pre-4.7 Tokenizer
Opus 4.6 brought a 1M window, 128K output and adaptive thinking to Opus at $5/$25. Why its older tokenizer and looser API still make it a sensible pick in late 2026.
Read
ModelsClaude Opus 4.7 Guide: The Release That Changed the API
Opus 4.7 kept the $5/$25 price but changed the tokenizer, removed thinking budgets and added xhigh effort. What changed, what it costs and whether to still run it.
Read
ModelsClaude Opus 4.8 Guide: The Opus 4 Model That Outlived Its Successors
Opus 4.8 shipped as a modest upgrade, then became the fallback newer Opus models route cyber requests to. What it changed, what it scores and when to choose it.
ReadModelsClaude Opus 5 Guide: Refusals, Default Thinking and When to Stay
Opus 5 put thinking on by default and added a refusal stop reason your code must handle. What changed from Opus 4.8, and why some teams should not move to 5.5 yet.
ReadModelsClaude Opus 5.5 Guide: Cheaper Than Opus 5, Close to Fable 5.1
Opus 5.5 cut Opus pricing to $4/$20 and beat Fable 5.1 on Anthropic's own table. Benchmarks, the medium effort default, four breaking changes and how to test it.
ReadModelsClaude Sonnet 4 After Retirement: Where It Runs and How to Move
Anthropic retired Claude Sonnet 4 on its own API in June 2026. What the model was, where it still runs, and what breaks when you migrate to Sonnet 5.5.
ReadModelsClaude Sonnet 4.5 Guide: Retirement Date and Where to Migrate
Sonnet 4.5 is deprecated and retires on the Claude API on 30 November 2026. What it was, what breaks when you move, and whether to land on Sonnet 4.6 or Sonnet 5.5.
ReadShowing 13–24 of 731 articles