Anthropic's Claude models on this feed are hosted APIs with text, image, and file input. The practical ladder is Fable for the hardest technical work, Sonnet for everyday coding and writing, and Haiku for high-volume drafts. Opus sits between Fable and Sonnet on price.
Flagship · $8.00 in · $40.00 out · 1,000,000 tokens
Guide note · 19 Nov 2026
OpenRouter calls Claude Fable 5.2 Anthropic's most capable model for demanding technical work, with gains in coding, reasoning, and vision over Fable 5.1. It is the model to spend money on when the problem is actually hard.
Paste the timeline, the logs, and a screenshot of the dashboard. Ask for hypotheses you can falsify, not for a heroic root cause. The listing is aimed at this kind of technical work.
Claude Fable 5.2 is listed at $8.00 input and $40.00 output per million tokens, with a 1,000,000-token context window.
Read a dense contract for flags, not for legal advice
File input and a long window can hold the agreement. Ask for dates, parties, and clauses that shift liability. A person who owns the decision still reads those clauses.
Claude Fable 5.2 is listed at $8.00 input and $40.00 output per million tokens, with a 1,000,000-token context window.
The tasks are editorial illustrations, not measured benchmarks. The prices in them are read from today's feed.
Everyday · $2.00 in · $10.00 out · 1,000,000 tokens
Guide note · 19 Nov 2026
OpenRouter describes Claude Sonnet 5.5 as the Sonnet-class model for well-scoped everyday work, especially building features, fixing bugs, and producing finished text. It is the reviewed default for most coding and writing.
A written spec for a small app — accounts, a form, a list — matches the listing's 'build the feature, fix the bug' brief. Keep the spec shorter than your ambition.
Claude Sonnet 5.5 is listed at $2.00 input and $10.00 output per million tokens, with a 1,000,000-token context window.
Edit a 2,000-word essay for structure
Ask for a reverse outline and the three weakest paragraphs. Everyday writing is the job. The long window means you can include the brief and the draft together.
Claude Sonnet 5.5 is listed at $2.00 input and $10.00 output per million tokens, with a 1,000,000-token context window.
The tasks are editorial illustrations, not measured benchmarks. The prices in them are read from today's feed.
OpenRouter describes Claude Haiku 5.5 as Anthropic's small, fast model for high-volume work such as summarisation, subagents, and browser use. Stronger than Haiku 4.5 at coding, on the listing's own words, and far cheaper than Sonnet.
Route the message and draft two sentences. The listing names high-volume summarisation and subagents. Sample a hundred outputs before you trust the queue.
Claude Haiku 5.5 is listed at $0.10 input and $0.50 output per million tokens, with a 1,000,000-token context window.
Summarise a meeting transcript you paste in
A one-hour transcript is a summarisation job, which the listing names, and it fits the million-token window easily.
Claude Haiku 5.5 is listed at $0.10 input and $0.50 output per million tokens, with a 1,000,000-token context window.
The tasks are editorial illustrations, not measured benchmarks. The prices in them are read from today's feed.
For most people, start with Claude Fable 5.2, then drop to the next model on this page when the task is routine. Read the cheaper-model note under each example.
Are the examples benchmarks?
No. They are concrete tasks used to explain fit. The prices next to them are live figures from the feed. The judgment is a guide note.