Best LLMs for Writing in 2026
As of October 6, 2026, the top model in best llms for writing on the BenchLM leaderboard is MAI-Thinking-1 with a score of 94.7.
Bottom line: instruction following is what separates writing models in practice. Claude Fable 5 held the top human-preference rating in the September 2026 snapshot; the instruction-following leaders below execute a brief most faithfully.
This ranking moves when models ship. Get the releases, price changes and retirements that affect your shortlist. Follow model changes
Full Rankings (124 ranked)
Current position
MAI-Thinking-1 leads the live llms for writing ranking at 94.7.
Grok 4.3 ranks #2 at 93.2.
GPT-5.2-Codex ranks #3 at 92.4.
How to choose
Key Takeaways
The top model is MAI-Thinking-1 by Microsoft with a provisional score of 94.7.
The best open-weight model is GLM-5.1 at position #4.
124 models are included in this ranking.
Score in Context
What these scores mean
Writing quality has no direct benchmark, so this page ranks by instruction following — whether the model does what the brief asked. Read it with a blind human-preference rating from an independent index for prose quality.
Known limitations
Style is subjective and prompt-sensitive. Instruction-following scores reward constraint compliance, not voice — a model can follow your brief perfectly and still write flat prose. Test your actual editing workflow.
About this ranking
Ranking data as of October 6, 2026
There is no single "writing benchmark," so BenchLM ranks writing capability by the instruction-following category — the best available proxy for whether a model matches your tone, structure, and length constraints — read alongside human-preference ratings from an independent index. Claude models held the top of that index in the September 2026 snapshot; the instruction-following table shows who follows a brief most reliably.
Unless noted otherwise, ranking surfaces on this page use BenchLM’s provisional leaderboard lane rather than the stricter sourced-only verified leaderboard.
MAI-Thinking-1 leads this ranking with a score of 94.7, followed by Grok 4.3 (93.2) and GPT-5.2-Codex (92.4). The top three are separated by just a few points — any of them would perform well for this use case.
The best open-weight option is GLM-5.1 (ranked #4 with a score of 92.4). While proprietary models lead, open-weight options are within striking distance for teams willing to trade a few points of performance for full model control.
This ranking uses provisional weighted averages across the scoring benchmarks in instructionFollowing. For detailed model profiles, click any model name above. To compare two specific models head-to-head, use the "vs #" links.
Questions
What is the best LLM for writing?
For prose, our pick is Claude Fable 5: it held the highest human-preference rating we track in the September 2026 snapshot, the best available signal for how writing reads. For structured copy with strict format and length rules, start instead from the leaders of the instruction-following table above, which recomputes on every build.
What is the best LLM for creative writing?
Human preference is the best available signal for creative prose, and Claude models led the September 2026 snapshot of the independent index we track, with Claude Fable 5 on top. Non-reasoning models often feel more natural for iterative drafting because they respond without a thinking pause.
What is the best AI for resume writing?
Resume writing is an instruction-following task: strict format, tight length, specific tone. Any model near the top of the table above handles it well, so the practical differences are price and speed. Pick the cheapest row that clears about 90 on the price-performance page, then test it on your own template.
Are benchmarks meaningful for writing quality?
Partially. Instruction following measures whether the model obeyed the brief — essential for professional writing — and an independent human-preference rating captures blind reader preference. Neither measures your voice. Use the scores to shortlist, then run a 10-prompt bake-off in your own editing workflow.
Explore More
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 5,500+ readers.
One email each week. Unsubscribe anytime.