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.
- Claude vs GeminiClaude Opus 5.5 (High)1532.1 vs 1524.4 · Δ8vs Gemini 4 Argon (High)
- Cheapest in top 20
Gemini 3.7 Flash (High)$1.5 / 1MScore 1491.5 · #12 - Best score per $ in top 20Gemini 3.7 Flash (High)994 / $Score 1491.5 · $1.5
Claude vs Gemini: which is better for writing?
| Model | Score | Rank | Creative writing | Input / output | Score per $ |
|---|---|---|---|---|---|
| Claude Opus 5.5 (High)Top Claude | 1532.1±11.9 | 1 | 1534.2 #1 | $4 / $20 | 192 |
| Claude Sonnet 5.5 (xHigh)Cheapest Claude in top 20 | 1475.7±12.5 | 18 | 1468.3 #18 | $2 / $10 | 369 |
| Gemini 4 Argon (High)Top Gemini | 1524.4±11.9 | 2 | 1523.1 #2 | $2 / $10 | 381 |
| Gemini 3.7 Flash (High)Cheapest Gemini in top 20 | 1491.5±7.5 | 12 | 1489.1 #9 | $0.75 / $3.75 | 994 |
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 · Writing & literature
417 modelsSorted 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 | Creative writing | Input / output | Blended | Score per $ |
|---|---|---|---|---|---|---|---|---|
| 1 |  LLaMA 13BMeta |
809±27.7 | 417 | 513 | 793.7 | No price yet | ||
| 2 | ?StableLM Tuned Alpha 7B | 867.9±23.4 | 416 | 681 | 867.6 | No price yet | ||
| 3 | ?Dolly 2.0 12B | 872.2±22.9 | 415 | 732 | 863.5 | No price yet | ||
| 4 | ?FastChat-T5 3B | 907.5±20.6 | 414 | 940 | 901.9 | No price yet | ||
| 5 | ?ChatGLM 6B | 925.3±19.9 | 413 | 1,056 | 899.8 | No price yet | ||
| 6 | ?OpenAssistant Pythia 12B | 928.2±17.5 | 412 | 1,408 | 923.1 | No price yet | ||
| 7 | ?RWKV-4 Raven 14B | 934±19 | 411 | 1,098 | 933.5 | No price yet | ||
| 8 | ?ChatGLM2 6B | 952.7±22.5 | 410 | 775 | 940.9 | No price yet | ||
| 9 | ?Alpaca 13B | 968.2±19.2 | 409 | 1,217 | 967.3 | No price yet | ||
| 10 | ?ChatGLM3 6B | 972.5±20.1 | 408 | 1,108 | 969.8 | No price yet | ||
| 11 | ?MPT 7B Chat | 972.8±20.4 | 407 | 892 | 957.4 | No price yet | ||
| 12 | ?GPT4All 13B Snoozy | 973.1±31.4 | 406 | 347 | 971.8 | No price yet | ||
| 13 | Gemma 2B ITGoogle |
978.3±18.5 | 405 | 1,152 | 986.2 | No price yet | ||
| 14 | ?Koala 13B | 979.4±17.6 | 404 | 1,515 | 972.7 | No price yet | ||
| 15 |  Qwen1.5-4B-ChatAlibaba |
1000.8±14.8 | 403 | 1,895 | 969 | No price yet | ||
| 16 |  OLMo 7B InstructAi2 |
1003.1±17.8 | 402 | 1,511 | 989.5 | No price yet | ||
| 17 | Gemma 1.1 2B ITGoogle |
1006±13.1 | 401 | 2,596 | 997.7 | No price yet | ||
| 18 | PaLM 2Google |
1006.8±15 | 400 | 2,155 | 991.2 | No price yet | ||
| 19 | Code Llama 70B InstructMeta |
1014.8±36.5 | 399 | 262 | – | No price yet | ||
| 20 | ?SmolLM2 1.7B Instruct | 1024.5±26.1 | 398 | 627 | 1031.2 | No price yet | ||
| 21 | Llama 3.2 1B InstructMeta |
1029.7±14.7 | 397 | 2,047 | 1032.5 | No price yet | ||
| 22 |  Mistral 7B InstructMistral AI |
1030.3±15 | 396 | 2,336 | 1028.1 | No price yet | ||
| 23 | Gemma 7B ITGoogle |
1033.8±14.6 | 395 | 2,191 | 1023.3 | No price yet | ||
| 24 | ?StripedHyena Nous 7B | 1034.6±17.8 | 394 | 1,271 | 1029 | No price yet | ||
| 25 | Llama 2 7B ChatMeta |
1037.1±11.5 | 393 | 3,454 | 1033 | No price yet | ||
| 26 |  Phi-3 Mini 128K InstructMicrosoft |
1040.3±11.6 | 392 | 4,627 | 1024 | No price yet | ||
| 27 | Code Llama 34B InstructMeta |
1048.3±14.5 | 391 | 2,003 | 1031.4 | No price yet | ||
| 28 | ?Vicuna 7B | 1048.5±15 | 390 | 1,783 | 1020.8 | No price yet | ||
| 29 | Qwen-14B-ChatAlibaba |
1049.9±18.1 | 389 | 1,230 | 1027.4 | No price yet | ||
| 30 |  Granite 3.0 2B InstructIBM |
1050.5±15.1 | 388 | 1,820 | 1046 | No price yet | ||
| 31 | Phi-3 Mini 4K InstructMicrosoft |
1054.4±10.3 | 387 | 5,207 | 1036.7 | No price yet | ||
| 32 | Phi-3 Mini 4K Instruct (June 2024)Microsoft |
1056.6±11.2 | 386 | 3,382 | 1044.4 | No price yet | ||
| 33 | Llama 2 13B ChatMeta |
1061.5±10.7 | 385 | 4,785 | 1046.6 | No price yet | ||
| 34 | Qwen1.5-7B-ChatAlibaba |
1064.3±17.7 | 384 | 1,093 | 1034.3 | No price yet | ||
| 35 | ?Guanaco 33B | 1066.3±22.7 | 383 | 714 | 1079.7 | No price yet | ||
| 36 | Mistral 7B Instruct v0.2Mistral AI |
1073±10.5 | 382 | 4,703 | 1067.4 | No price yet | ||
| 37 | Gemma 1.1 7B ITGoogle |
1073.1±9.3 | 381 | 6,034 | 1059.2 | No price yet | ||
| 38 | Granite 3.0 8B InstructIBM |
1078.4±15.2 | 380 | 1,751 | 1070.4 | No price yet | ||
| 39 | ?Vicuna 13B | 1078.9±10.6 | 379 | 4,672 | 1057.9 | No price yet | ||
| 40 | ?MPT 30B Chat | 1079.3±22.5 | 378 | 647 | 1073.8 | No price yet | ||
| 41 | ?Zephyr 7B Alpha | 1081.6±25.2 | 377 | 569 | 1068.5 | No price yet | ||
| 42 | ?Zephyr 7B Beta | 1085.7±13.9 | 376 | 2,691 | 1102.1 | No price yet | ||
| 43 | ?Dolphin 2.2.1 Mistral 7B | 1085.9±28.4 | 375 | 382 | 1085.5 | No price yet | ||
| 44 |  SOLAR 10.7B Instruct v1.0Upstage |
1087.8±21 | 374 | 894 | 1089.1 | No price yet | ||
| 45 |  GPT-3.5 Turbo (1106)OpenAI |
1091±13.5 | 373 | 3,932 | 1034.8 | No price yet | ||
| 46 | Granite 3.1 2B InstructIBM |
1092±20.8 | 372 | 850 | 1115.7 | No price yet | ||
| 47 | Llama 2 70B ChatMeta |
1092.7±8.6 | 371 | 9,591 | 1074.6 | No price yet | ||
| 48 | Phi-3 Small 8K InstructMicrosoft |
1093.7±10 | 370 | 4,639 | 1082.6 | No price yet | ||
| 49 |  Llama 2 70B SteerLM ChatNVIDIA |
1094.4±21.3 | 369 | 836 | 1090.3 | No price yet | ||
| 50 | Llama 3.2 3B InstructMeta |
1095.5±13.8 | 368 | 2,118 | 1093.9 | No price yet | ||
Show all 417 models (showing top 50)
Data: LMArena (lmarena-ai), CC-BY-4.0 · Writing & literature · published Oct 22, 2026
Sources & method
- LMLMArena leaderboard dataset (Hugging Face, CC-BY-4.0)
config text · creative_writing · published Oct 22, 2026 · 416 models
- LMLMArena leaderboard dataset (Hugging Face, CC-BY-4.0)
config text · industry_writing_and_literature_and_language · published Oct 22, 2026 · 417 models
- LMLMArena leaderboard dataset (Hugging Face, CC-BY-4.0)
config text · longer_query · published Oct 22, 2026 · 396 models
- LMLMArena leaderboard dataset (Hugging Face, CC-BY-4.0)
config text · multi_turn · published Oct 22, 2026 · 416 models
- LMLMArena leaderboard dataset (Hugging Face, CC-BY-4.0)
config text · instruction_following · published Oct 22, 2026 · 418 models
- $
- Only the latest LMArena publish date is used; score, CI, votes and rank are copied from the dataset.
- Reasoning tiers of one model (e.g. High / Max) are separate LMArena entries and are listed separately.
- Prices are shown only when the entry maps to our LLM leaderboard; otherwise –, never estimated.
- Score ÷ blended price (3:1 input:output). A rough value indicator only.
- Synced daily at 06:35 UTC; if the source fails, the last good snapshot is kept.
FAQ
What is the best AI for writing?
In LMArena's creative writing category published Oct 22, 2026, #1 is Anthropic's Claude Opus 5.5 (High) (1534.2, 2,453 votes).
Claude or Gemini: which is better for writing?
Top creative-writing scores: Claude Claude Opus 5.5 (High) (1534.2, output $20/1M tokens); Gemini Gemini 4 Argon (High) (1523.1, 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 MiMo-V2.6-Pro (1466.4, #20) at about $0.544/1M tokens blended; the cheapest is MiMo-V2.6-Pro ($0.544/1M, 1466.4).
Which vendors rank highest for writing?
By the creative writing category, each vendor's best model ranks: Anthropic Claude Opus 5.5 (High) (1534.2, #1), Google Gemini 4 Argon (High) (1523.1, #2), Meta Muse Wonder (Max) (1492.6, #7), Alibaba Qwen3.8-Max (1477.4, #16), Xiaomi MiMo-V2.6-Pro (1466.4, #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.