Recommendation for Creative & fiction

Creative Writing

Our top recommendation for Creative Writing, based on the public evidence we track, is Anthropic: Claude Fable 5.[1] Google: Gemini 3.6 Flash is the next-ranked alternative. Use for creative writing when you need a model in the top tier of human preference rankings without requiring the absolute highest placement.

About this recommendation

Updated
Sep 4, 2026
Evidence through
Sep 4, 2026
Sources
4
Revision
v56

Decision audit

Why this result

Inspect the inputs and the computed order behind the recommendation.

Models screened

21

live candidates

Evaluation feeds

5

task-weighted

Winner coverage

100%

intended feed weight

Largest provider share

2 of 4

Anthropic

Provisional source breadth. 2 citation families and 1 practitioner families support the top result; 0 cautionary threads is retained. The largest citation family contributes 80%.

Sources evaluated

The task sets these weights before any model is scored.

winner: Claude Fable 5
Evaluation feedWeightWinner resultField measured
LMArena Creative Writing
30%
#219/21
LiveBench Instruction Following
25%
#521/21
LMArena Text
20%
#119/21
LiveBench Language
15%
#121/21
OpenRouter usage
10%
87/10021/21

Provider concentration

Each exact model is scored separately; provider identity is not a ranking input.

Anthropic50%
  • Anthropic2 models
  • Google1 model
  • Z.ai1 model

Decision table

Every published model is shown in computed order. Practitioner sources are distinct community threads, not the citations repeated in the prose below.

RankModelRelative scoreCoveragePractitioner evidenceStrongest measured reason
01Claude Fable 5Anthropic
94
100%2 threads · 1 families · 0 cautions#1 LiveBench Language · #1 LMArena Text
02Gemini 3.6 FlashGoogle
80
100%no linked practitioner threads#7 LiveBench Instruction Following · #10 LiveBench Language
03Claude Opus 4.6Anthropic
80
100%no linked practitioner threads#1 LMArena Creative Writing · #2 LMArena Text
04GLM 5.2Z.ai
74
100%1 threads · 1 families · 1 cautions#10 LMArena Creative Writing · #20 LMArena Text

Relative score combines normalized benchmark quality and signal coverage; independent practitioner evidence and freshness are bounded tie-breakers. It is an ordering score, not an absolute quality percentage. The writing model receives this order and cannot change it.

  1. Anthropic: Claude Fable 5 ranks #2 of 144 on LMArena's creative-writing category (Elo 1497), based on blind human preference votes.

    Best when: Consider only after reviewing the cited caution.

  2. Places in the top ten for creative-writing preference, showing competitive prose generation capabilities.

    Best when: Use for creative writing when you need a model in the top tier of human preference rankings without requiring the absolute highest placement.

    Tips

    • Use for creative writing when you need a model in the top tier of human preference rankings without requiring the absolute highest placement.
      Source 1
      Ranks #10 of 144 on LMArena's creative-writing category (Elo 1463), based on blind human preference votes.
      LMArena creative-writing categoryOpen original ↗
  3. Leads the creative-writing category on LMArena with the highest Elo score, indicating strong human preference for its prose quality and imaginative output.

    Best when: Use when you need fiction or poetry that human evaluators consistently prefer over 143 competing models.

    Tips

    • Use when you need fiction or poetry that human evaluators consistently prefer over 143 competing models.
      Source 2
      Ranks #1 of 144 on LMArena's creative-writing category (Elo 1505), based on blind human preference votes.
      LMArena creative-writing categoryOpen original ↗
  4. An open-weight model with strong overall text arena performance, though its creative-writing specific ranking is not directly evidenced; community reports suggest utility for hard-science fiction development through debate workflows.

    Best when: Use for hard-science fiction development where you need models to debate physics accuracy, as one writer used it alongside Gemini and Claude to catch errors in technical story plots.

    Tips

    • Use for hard-science fiction development where you need models to debate physics accuracy, as one writer used it alongside Gemini and Claude to catch errors in technical story plots.
      Source 3
      > it is clear that actual intelligence has plateaued significantly N=1, but I disagree strongly. I'm writing a hard-science science fiction story, and the physics of it is at (and frankly, beyond) my skillset. The story's plot has had to change over a dozen times as I realized errors in my application of physics in the story. Throughout, I've been reviewing the physics with LLMs, mainly Gemini 3.1 Pro Preview, but also with Claude and OpenAI. Often I have the LLMs debate each other -- "My frien…

    Watch out for

    • Watch for lack of direct creative-writing benchmark evidence, as its LMArena ranking is for overall text rather than the specific fiction and poetry category.
      Source 4
      Ranks #23 of 144 on LMArena's overall text arena (Elo 1472), based on blind human preference votes.
      LMArena text arenaOpen original ↗

Frequently asked

What is an alternative to Anthropic: Claude Fable 5?
Google: Gemini 3.6 Flash is the next-ranked option. Use for creative writing when you need a model in the top tier of human preference rankings without requiring the absolute highest placement.[1]

Sources

  1. 1

    Ranks #10 of 144 on LMArena's creative-writing category (Elo 1463), based on blind human preference votes.

    LMArena creative-writing category · Benchmark · Sep 1, 2026
  2. 2

    Ranks #1 of 144 on LMArena's creative-writing category (Elo 1505), based on blind human preference votes.

    LMArena creative-writing category · Benchmark · Sep 1, 2026
  3. 3

    > it is clear that actual intelligence has plateaued significantly N=1, but I disagree strongly. I'm writing a hard-science science fiction story, and the physics of it is at (and frankly, beyond) my skillset. The story's plot has had to change over a dozen times as I realized errors in my application of physics in the story. Throughout, I've been reviewing the physics with LLMs, mainly Gemini 3.1 Pro Preview, but also with Claude and OpenAI. Often I have the LLMs debate each other -- "My frien…

    gcanyon · Hacker News · Jun 20, 2026
  4. 4

    Ranks #23 of 144 on LMArena's overall text arena (Elo 1472), based on blind human preference votes.

    LMArena text arena · Benchmark · Sep 1, 2026

Rankings synthesized from community evidence and open benchmarks. See methodology. Not driven by vendor marketing.