Best LLM for Creative Writing: AI Writing Leaderboard

The writing board covers five writing-related categories of the LMArena text leaderboard, one per tab: creative writing (creative_writing, default), writing/literature/language tasks, longer prompts (longer_query; not a long-context test), multi-turn (multi_turn) and instruction following (instruction_following). Scores come from blind human pairwise votes, not automated benchmarks.

#1 in Creative writingData Nov 8, 2026
Gemini 4 Argon (High)Google · 1535.8

Bottom line (Nov 8, 2026 data): the best LLM for writing is Gemini 4 Argon (High) (1535.8). Claude Opus 5.5 (High) (1531.2) leads Gemini 4 Argon (High) (1535.8) by 4 points, but Gemini's output price is 50% of Claude's. Best value: Gemini 3.7 Flash (High) (highest score per dollar, $1.5/1M tokens blended).

Claude vs Gemini: which is better for writing?

ModelScoreRankWriting & literatureInput / outputScore per $
Claude Opus 5.5 (High)Top Claude1531.2±11.631528.8 #3$4 / $20191
Gemini 4 Argon (High)Top Gemini1535.8±10.411534.7 #2$2 / $10384
Gemini 3.7 Flash (High)Cheapest Gemini in top 201485±8.2121488.9 #13$0.75 / $3.75990

When the gap is smaller than the combined confidence intervals, the rank difference may not be significant. Prices are OpenRouter quotes (USD / 1M tokens).

Full ranking · Creative writing

418 models

Sorted by LMArena score; ± is the 95% CI half-width. Prices: OpenRouter quote (official list price if none), USD per 1M tokens.

# Model Score Rank Votes Writing & literature Input / output Blended Score per $
1 Gemini 4 Argon (High)Google 1535.8±10.4 1 4,517 1534.7 $2 / $10 $4 384
2 Gemini 4 Carbon (High)Google 1534.9±20 2 1,011 1544.3 $4 / $20 $8 192
3 Claude Opus 5.5 (High)Anthropic 1531.2±11.6 3 3,562 1528.8 $4 / $20 $8 191
4 Claude Fable 5.1 (Max)Anthropic 1513.4±10.1 4 4,516 1514.1 $10 / $50 $20 76
5 Claude Opus 4.6 (High)Anthropic 1503.8±6.1 5 15,614 1501.8 $5 / $25 $10 150
6 Muse Wonder (Max)Meta 1494.9±12.7 6 2,659 1498.7 $4 / $20 $8 187
7 Claude Fable 5 (High)Anthropic 1494.4±7.2 7 9,853 1502.1 $10 / $50 $20 75
8 Claude Opus 5 (Max)Anthropic 1493.8±7.9 8 8,175 1498.4 $5 / $25 $10 149
9 Gemini 3.8 Flash (High)Google 1493.6±6.8 9 13,544 1497.2 $0.825 / $4.125 $1.65 905
10 Claude Opus 5 (High)Anthropic 1486.4±6.4 10 16,555 1494.1 $5 / $25 $10 149
11 Claude Opus 4.7 (High)Anthropic 1486.1±6.9 11 11,583 1492.8 $5 / $25 $10 149
12 Gemini 3.7 Flash (High)Google 1485±8.2 12 7,412 1488.9 $0.75 / $3.75 $1.5 990
13 Claude Opus 4.6Anthropic 1483.5±5.9 13 15,842 1491.7 $5 / $25 $10 148
14 Gemini 3.1 Pro PreviewGoogle 1482.9±5.2 14 24,670 1481.3 $2 / $12 $4.5 330
15 Qwen3.8-MaxAlibaba 1482±8.1 15 7,308 1478.8 $2 / $6 $3 494
16 Gemini 3 ProGoogle 1481.8±8.1 16 6,488 1477.7 $2 / $12 $4.5 329
17 Claude Opus 4.7Anthropic 1479.7±6.9 17 11,820 1479.1 $5 / $25 $10 148
18 Gemini 3.5 Flash (Medium)Google 1470.5±7.4 18 9,264 1472.9 $1.5 / $9 $3.375 436
19 Gemini 3.5 Flash (High)Google 1468.9±7.2 19 9,271 1478.6 $1.5 / $9 $3.375 435
20 MiMo-V2.6-ProXiaomi 1467.8±21.1 20 801 1469.7 No price yet
21 Claude Sonnet 5.5 (xHigh)Anthropic 1467.5±12 21 3,119 1470.6 $2 / $10 $4 367
22 Muse Spark 1.3 (Max)Meta 1466.3±9.1 22 5,674 1474.5 $1.25 / $4.25 $2 733
23 Gemini 3.6 Flash (High)Google 1466.1±8 23 7,368 1470.3 $0.75 / $3.75 $1.5 977
24 Qwen3.5-Max-PreviewAlibaba 1465.3±10.8 24 3,280 1466.9 No price yet
25 GLM-5.3 (Max)Z.ai 1460±8.3 25 6,967 1469.5 $0.07 / $1.4 $0.402 3,628
26 GLM-5.2 (Max)Z.ai 1459.3±6.8 26 10,866 1461.2 $0.006 / $6.8 $1.705 856
27 Muse SparkMeta 1458.9±13.7 27 2,003 1454.6 No price yet
28 Gemini 3 FlashGoogle 1456.1±9.2 28 4,781 1458.7 $0.5 / $3 $1.125 1,294
29 Claude Opus 4.8 (High)Anthropic 1455.1±6.5 29 12,548 1462.2 $5 / $25 $10 146
30 Gemini 2.5 ProGoogle 1454±5.1 30 17,594 1454.1 $1.25 / $10 $3.438 423
31 GPT-5.5 (High)OpenAI 1453.8±6.5 31 12,899 1465.3 $5 / $30 $11.25 129
32 GLM-5.1Z.ai 1453.3±6.2 32 13,171 1453.8 $1.4 / $4.4 $2.15 676
33 Kimi K3 (Max)Moonshot AI 1453.2±7.6 33 8,344 1467.9 $0.192 / $16 $4.144 351
34 GPT-5.5OpenAI 1451.7±6.5 34 13,170 1463.2 $5 / $30 $11.25 129
35 GPT-5.6 Sol (xHigh)OpenAI 1450.8±7.4 35 9,307 1464 $4 / $20 $8 181
36 Qwen3.7-Max-PreviewAlibaba 1447.8±26.9 36 501 1463.2 $1.475 / $4.425 $2.213 654
37 Claude Opus 4.8Anthropic 1447.1±6.5 37 12,584 1453.1 $5 / $25 $10 145
38 DeepSeek-V4-ProDeepSeek 1446.6±7.2 38 9,582 1447.1 $0.955 / $1.911 $1.194 1,212
39 Claude Opus 4.5 (High, 32K, 20251101)Anthropic 1444.5±8.3 39 5,630 1444.4 $5 / $25 $10 145
40 GLM-5.3-FlashZ.ai 1442.7±7.1 40 10,620 1456.8 $0.034 / $0.083 $0.046 31,431
41 Claude Opus 4.5 (20251101)Anthropic 1442.7±6.4 41 11,287 1451.2 $5 / $25 $10 144
42 Claude Sonnet 4.5 (20250929)Anthropic 1442.2±6 42 12,300 1442 No price yet
43 Grok 4.5xAI 1442.1±7.7 43 8,086 1445.1 $2 / $6 $3 481
44 DeepSeek-V4-Pro Preview (High)DeepSeek 1441.4±7.3 44 9,261 1437.6 No price yet
45 ERNIE 5.1Baidu 1440.7±8.3 45 6,355 1449.1 No price yet
46 Muse Spark 1.2 (xHigh)Meta 1439.6±11.7 46 3,145 1456.6 $1.25 / $4.25 $2 720
47 Grok 4.20 Beta 1xAI 1439.2±9.7 47 4,120 1428.1 No price yet
48 Qwen3.7-PlusAlibaba 1438.3±7.6 48 7,662 1437.3 $0.32 / $1.28 $0.56 2,568
49 GPT-5.4 (High)OpenAI 1437.8±6.9 49 10,966 1457.3 $2.5 / $15 $5.625 256
50 DeepSeek-V4.1-Flash-20260910 (Max)DeepSeek 1437.5±12 50 2,791 1449.4 $0.051 / $0.567 $0.18 7,986

Show all 418 models (showing top 50)

Data: LMArena (lmarena-ai), CC-BY-4.0 · Creative writing · published Nov 8, 2026

Sources & method

  1. Only the latest LMArena publish date is used; score, CI, votes and rank are copied from the dataset.
  2. Reasoning tiers of one model (e.g. High / Max) are separate LMArena entries and are listed separately.
  3. Prices are shown only when the entry maps to our LLM leaderboard; otherwise –, never estimated.
  4. Score ÷ blended price (3:1 input:output). A rough value indicator only.
  5. Synced daily at 06:35 UTC; if the source fails, the last good snapshot is kept.

This week

vs Nov 2, 2026

FAQ

What is the best AI for writing?

In LMArena's creative writing category published Nov 8, 2026, #1 is Google's Gemini 4 Argon (High) (1535.8, 4,517 votes).

Claude or Gemini: which is better for writing?

Top creative-writing scores: Claude Claude Opus 5.5 (High) (1531.2, output $20/1M tokens); Gemini Gemini 4 Argon (High) (1535.8, output $10/1M tokens).

What is the cheapest good LLM for writing?

Among the top 20 for creative writing, the best score per dollar is Gemini 3.7 Flash (High) (1485, #12) at about $1.5/1M tokens blended; the cheapest is Gemini 3.7 Flash (High) ($1.5/1M, 1485).

Which vendors rank highest for writing?

By the creative writing category, each vendor's best model ranks: Google Gemini 4 Argon (High) (1535.8, #1), Anthropic Claude Opus 5.5 (High) (1531.2, #3), Meta Muse Wonder (Max) (1494.9, #6), Alibaba Qwen3.8-Max (1482, #15), Xiaomi MiMo-V2.6-Pro (1467.8, #20).

Where do the scores come from?

From LMArena: users submit writing prompts, two anonymous models answer, users vote for the better one, and scores are fitted with a Bradley-Terry model.