Skip to content
Bencher

TestsBenchmarks · writing

LMArena Text - Creative Writing

Crowd-voted, style-controlled Elo on prompts classified as creative writing (stories, poems, scripts, original prose).

Elo ratingHigher is betterCounts toward:Weights: Writing & creativity ×1The test's websiteOfficial page ↗

The best 15 · 28 models tested

Top 15 · best setting per model · 28 models, 31 results (31 independent)

Measured by independent testersIndependent Reported by the makerVendor-reported
LMArena Text - Creative Writing leaderboard
Thinking levelSetting
1Gemini 4 ArgonGoogle (Gemini / DeepMind)1522Highgemini-4-argon-highIndependent testIndependentLMArena ↗—30 Sep 2026
2Claude Opus 5.5Anthropic1515Highclaude-opus-5.5-highIndependent testIndependentLMArena ↗—30 Sep 2026
3Claude Fable 5Anthropic1503Highclaude-fable-5-highIndependent testIndependentLMArena ↗—30 Sep 2026
4Claude Opus 4.6Anthropic1501Highclaude-opus-4-6-highIndependent testIndependentLMArena ↗—30 Sep 2026
5Gemini 3.7 FlashGoogle (Gemini / DeepMind)1496Highgemini-3.7-flash-highIndependent testIndependentLMArena ↗—30 Sep 2026
6Claude Opus 4.7Anthropic1489Highclaude-opus-4-7-highIndependent testIndependentLMArena ↗—30 Sep 2026
7Gemini 3 ProGoogle (Gemini / DeepMind)1484Defaultgemini-3-proIndependent testIndependentLMArena ↗—30 Sep 2026
8Gemini 3.8 FlashGoogle (Gemini / DeepMind)1484Highgemini-3.8-flash-highIndependent testIndependentLMArena ↗—30 Sep 2026
9Claude Fable 5.1Anthropic1482Maxclaude-fable-5.1-maxIndependent testIndependentLMArena ↗—30 Sep 2026
10Gemini 3.1 Pro (Preview)Google (Gemini / DeepMind)1480Defaultgemini-3.1-pro-previewIndependent testIndependentLMArena ↗—30 Sep 2026
11Claude Opus 5Anthropic1472Highclaude-opus-5-highIndependent testIndependentLMArena ↗—30 Sep 2026
12Gemini 3.6 FlashGoogle (Gemini / DeepMind)1472Highgemini-3.6-flash-highIndependent testIndependentLMArena ↗—30 Sep 2026
13Claude Opus 4.8Anthropic1469Highclaude-opus-4-8-highIndependent testIndependentLMArena ↗—30 Sep 2026
14Gemini 3.5 FlashGoogle (Gemini / DeepMind)1469Mediumgemini-3.5-flash-mediumIndependent testIndependentLMArena ↗—30 Sep 2026
15Claude Opus 4.5Anthropic1469Highclaude-opus-4-5-20251101-high-32kIndependent testIndependentLMArena ↗—30 Sep 2026
16GPT-5.6 SolOpenAI1469Extra highgpt-5.6-sol-xhighIndependent testIndependentLMArena ↗—30 Sep 2026
17Qwen3.8 Max (0803)Qwen (Alibaba)1468Defaultqwen3.8-maxIndependent testIndependentLMArena ↗—30 Sep 2026
18Kimi K3Moonshot AI (Kimi)1460Maxkimi-k3-maxIndependent testIndependentLMArena ↗—30 Sep 2026
19Muse Spark 1.3Meta (Meta Superintelligence Labs, Muse)1459Maxmuse-spark-1.3-maxIndependent testIndependentLMArena ↗—30 Sep 2026
20GLM-5.3Z.ai (Zhipu)1455Maxglm-5.3-maxIndependent testIndependentLMArena ↗—30 Sep 2026
21GPT-6 AstraOpenAI1448Maxgpt-6-astra-maxIndependent testIndependentLMArena ↗—30 Sep 2026
22MiMo-V2.6-ProXiaomi MiMo1445Defaultmimo-v2.6-proIndependent testIndependentLMArena ↗—30 Sep 2026
23DeepSeek V4 Pro (0813)DeepSeek1445Highdeepseek-v4-pro-high-20260813Independent testIndependentLMArena ↗—30 Sep 2026
24GPT-6 SolOpenAI1438Maxgpt-6-sol-maxIndependent testIndependentLMArena ↗—30 Sep 2026
25DeepSeek V4.1 FlashDeepSeek1438Maxdeepseek-v4.1-flash-maxIndependent testIndependentLMArena ↗—30 Sep 2026
26Claude Sonnet 5Anthropic1436Highclaude-sonnet-5-highIndependent testIndependentLMArena ↗—30 Sep 2026
27Grok 4.7SpaceXAI (formerly xAI)1433Extra highgrok-4.7-xhighIndependent testIndependentLMArena ↗—30 Sep 2026
28GPT-6 LunaOpenAI1409Maxgpt-6-luna-maxIndependent testIndependentLMArena ↗—30 Sep 2026