Checked against the OpenRouter public models feed. Prices are copied, not estimated.

The Best Model ForWhich AI model to use, from ten labs.

The best AI model for writing

Draft or edit prose a person will still read.

Short answer

Start with Claude Sonnet 5.5 ($2.00 in · $10.00 out per million tokens). Sonnet is the reviewed pick for an essay or a careful email. Ask for a reverse outline before you ask for a rewrite, so you can see the structure it thinks you wrote. A cheaper stop is Claude Haiku 5.5 ($0.10 in · $0.50 out).

Quality

Claude Sonnet 5.5

Anthropic · $2.00 in · $10.00 out

Sonnet is the reviewed pick for an essay or a careful email. Ask for a reverse outline before you ask for a rewrite, so you can see the structure it thinks you wrote.

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.

Balanced

GPT-6.2 Sol

OpenAI · $2.00 in · $10.00 out

Sol is the everyday writer under Astra. One important email, or a first draft you will edit, fits it.

GPT-6.2 Sol is listed at $2.00 input and $10.00 output per million tokens, with a 1,050,000-token context window.

Cheapest sensible

GPT-6 Luna

OpenAI · $0.10 in · $0.50 out

Luna is the cheap reviewed writer for high volume. It is the wrong model for a letter you will send once to someone who is already angry.

GPT-6 Luna is listed at $0.10 input and $0.50 output per million tokens, with a 1,050,000-token context window.

Also on the feed

These match the factual limits of the task (modality, tools, context) and are not the reviewed picks. A row marked new has no guide note.

Questions

Will a flagship write better sentences?

Sometimes, and you will still edit. For most drafts the mid-tier models are enough. Spend flagship money on the draft that would be costly to get wrong.

Can the model match my house style?

Only if you paste examples. None of these listings include your style guide unless you supply it.