LLM Reference

The writing leaderboard · for knowledge workers

Best for writing

4 editor picks · 7 eligible models · Prose that doesn't read like a chatbot.

Editorial pick plus benchmark and API pricing context.

See raw /best
EDITOR'S CHOICEResearched 9d ago

Claude Opus 4.7

Anthropic · 1m context
Excellent

The best ear in the market — prose you'd actually ship.

Tops Chatbot Arena (1503) and writes paragraphs you'd ship; understands tone notes and edits like a copy chief.

Open model
The numbers
$/1M out
$25.00
$5.00 input
Context
1m
max window
Pros
  • +#1 Chatbot Arena
  • +Best voice & tone control
  • +Strong long-form structure
Cons
  • −$25 / 1M out

Also worth picking

The runners-up

ranked by editorial pick orderEditorial tiersExcellentStrongSolid
#2
GPT-5.5
OpenAI · 1.05m
$30.00 / 1M out
Arena 1488 — more predictable structure, slightly less voice; great for templated long-form.
#3
Gemini 3 Pro
Google DeepMind · 1m
$5.00 / 1M out
Arena 1486 with a 1M window — best for long drafts you'll polish anyway.
#4
Claude Sonnet 4.6
Anthropic · 1m
$15.00 / 1M out
Most of the Opus ear at $15 out — the everyday writing workhorse.

Eligibility

7 models are eligible for this board

Eligibility means tagged with useCases: [writing]. Pins must come from this pool.

All picks