The best AI for creative writing in 2026

There is no single best model for creative writing, because "creative writing" covers a four-line poem, a wedding speech and a ninety-thousand-word novel. So this page gives you our test results and says plainly which model is strongest at what, including where our own models are not the ones to pick. The story and EQ scores below are Hemmingway-1's, our first model. Our newest, Hemmingway-1.5, released on 25 October 2026 and built on it, was tested on AI detectors instead: 94% of its stories and speeches read as human on the first try.

Long stories: StoryBench

StoryBench is our own story test. Each model writes stories to the same prompts, and every story goes head to head with another model's in a blind judgement, run in both orders. Scores are on an Elo scale: higher wins more often.

For long fiction, the big story models win. Fable 5 Max and GLM-5.3 write better long stories than Hemmingway-1 in our tests, and so do the newest frontier models. Hemmingway-1 sits level with Kimi K3 and well ahead of Qwen3.8 Max and DeepSeek V4 Pro, and 504 points above the model it was built from. For a 27B model that is a lot, but it is not first.

Reading people: EQ-Bench 4

A lot of creative writing is getting a character's feelings right. EQ-Bench 4 is a public emotional-intelligence benchmark, and we ran it with its official harness.

Hemmingway-1 is third, inside twelve points of the top.

Short pieces that have to sound like a person

This is where Hemmingway is strongest: letters, speeches, messages, short stories and poems that need to read as if someone sat down and wrote them. In our Human-Likeness test, where a judge picks which of two texts a person wrote, Hemmingway-1 came first, twenty-six points ahead of the next model. Hemmingway-1.5, the model the app now writes with, goes further: GPTZero & Pangram and four of the top open AI detectors read 94% of its stories and 94% of its speeches as human, first try. How that was tested.

Which to pick

• A novel, or long chapters: one of the large story models. Fable 5 Max did best in our tests.
• Letters, speeches, cards, short stories, poems: Hemmingway, which is built to sound like a person and keeps to the length you asked for.
• Something you will run yourself: Hemmingway-1.5 and Hemmingway-1 are open weights under CC BY-NC 4.0, free for non-commercial use. More on open models for writing.

About these numbers

StoryBench and Human-Likeness are our own benchmarks, and we say so on Hemmingway-1's model card. The detector test is ours too, on Hemmingway-1.5's. Every matchup is blind, run in both orders, and judged by a model that is not one of the ones being judged. EQ-Bench 4 is not ours.

A sample

Here is a story Hemmingway-1 wrote, unedited, for the prompt "Write a story of about 150 words about a lighthouse keeper who gets a letter meant for someone else."

The supply boat came up the rock in November rain, and the boy who brought the stores had one more thing: a letter, addressed to Mr. Brannan, keeper of Vane Rock Light.

Callum took it without opening it, as he always did. After the lamp was lit and the log signed, he broke the seal. It was from a woman in Oban, Brannan's sister. She said she was leaving the mainland in spring. She said the things she had not said in the autumn were said now, all of them, and that she did not expect an answer.

Callum had kept letters like this for thirty years, for other men, other lights. None had ever been sent on. He read hers twice, then folded it along its old creases.

At four bells he climbed the stairs and trimmed the wick. The beam went out over the water, and the letter stayed in the drawer.

Try HemmingwayDownload the app

Common questions

What is the best AI for writing a novel?

In our StoryBench, Fable 5 Max scored highest at 1277, followed by GLM-5.3 at 1254. For a novel or long chapters, one of the large story models is the better pick. Hemmingway-1 is not first here: it sits level with Kimi K3 at 1197.

What is the best AI for short stories, poems and letters?

For short pieces that have to read as if a person wrote them, Hemmingway is built for the job. AI detectors read 94% of Hemmingway-1.5's stories as human, first try, and in our Human-Likeness test Hemmingway-1 came first, twenty-six points ahead of the next model. Both tests are our own.

Is there a good open-source AI for creative writing?

Hemmingway-1.5 and Hemmingway-1 are 27B models with open weights under CC BY-NC 4.0, free for non-commercial use. On our StoryBench Hemmingway-1 scored 1197, level with Kimi K3 and 504 points above Qwen3.8 27B, the model it was built from. More on open models for writing.

Which AI model has the best emotional intelligence?

On the public EQ-Bench 4, which we ran with its official harness, Fable 5 was first with 1341. Kimi K3 was second with 1332 and Hemmingway-1 third with 1330, ahead of GPT-5.5, Opus 4.7 and Opus 4.8.