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.
"Longer prompts" is LMArena's longer_query category: only battles where the user's prompt itself is long. It reflects handling of detailed, complex requests; it is not a long-context (hundreds of thousands of tokens) test.
- Claude vs GeminiGemini 4 Argon (High)1532.8 vs 1537.8 · Δ5vs Claude Opus 5.5 (High)
- Cheapest in top 20
MiMo-V2.6-Pro$0.544 / 1MScore 1500.5 · #11 - Best score per $ in top 20MiMo-V2.6-Pro2,759 / $Score 1500.5 · $0.544
Claude vs Gemini: which is better for writing?
| Model | Score | Rank | Creative writing | Input / output | Score per $ |
|---|---|---|---|---|---|
 Claude Opus 5.5 (High)Top Claude | 1532.8±10 | 2 | 1534.2 #1 | $4 / $20 | 192 |
| Claude Sonnet 5.5 (xHigh)Cheapest Claude in top 20 | 1494.7±10.4 | 13 | 1468.3 #18 | $2 / $10 | 374 |
| Gemini 4 Argon (High)Top Gemini | 1537.8±9.8 | 1 | 1523.1 #2 | $2 / $10 | 385 |
| Gemini 3.7 Flash (High)Cheapest Gemini in top 20 | 1494.3±6.4 | 14 | 1489.1 #9 | $0.75 / $3.75 | 996 |
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 · Longer prompts
396 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 | ?ChatGLM3 6B | 951.5±39.2 | 396 | 205 | 969.8 | No price yet | ||
| 2 | Gemma 2B ITGoogle |
980.3±29.5 | 395 | 377 | 986.2 | No price yet | ||
| 3 | ?StripedHyena Nous 7B | 982.2±31.2 | 394 | 334 | 1029 | No price yet | ||
| 4 |  Qwen1.5-4B-ChatAlibaba |
987.4±23.5 | 393 | 692 | 969 | No price yet | ||
| 5 |  Phi-3 Mini 128K InstructMicrosoft |
995.4±15.8 | 392 | 2,074 | 1024 | No price yet | ||
| 6 |  Llama 2 70B SteerLM ChatNVIDIA |
997.5±40.6 | 391 | 208 | 1090.3 | No price yet | ||
| 7 |  Llama 2 7B ChatMeta |
998.7±19.8 | 390 | 835 | 1033 | No price yet | ||
| 8 | ?Snowflake Arctic Instruct | 999±14.2 | 389 | 2,793 | 1080.7 | No price yet | ||
| 9 |  Mistral 7B InstructMistral AI |
1000.3±26.5 | 388 | 448 | 1028.1 | No price yet | ||
| 10 | Gemma 1.1 2B ITGoogle |
1002.3±17.7 | 387 | 1,179 | 997.7 | No price yet | ||
| 11 | PaLM 2Google |
1009.4±36.4 | 386 | 251 | 991.2 | No price yet | ||
| 12 | Code Llama 34B InstructMeta |
1013.1±32.5 | 385 | 320 | 1031.4 | No price yet | ||
| 13 | Gemma 7B ITGoogle |
1021.4±22.2 | 384 | 676 | 1023.3 | No price yet | ||
| 14 | Qwen-14B-ChatAlibaba |
1028.2±37.9 | 383 | 214 | 1027.4 | No price yet | ||
| 15 | ?Vicuna 33B | 1028.3±18.9 | 382 | 1,047 | 1124.4 | No price yet | ||
| 16 | ?Vicuna 7B | 1035.2±42.8 | 381 | 162 | 1020.8 | No price yet | ||
| 17 | Phi-3 Mini 4K Instruct (June 2024)Microsoft |
1041.2±18.6 | 380 | 1,035 | 1044.4 | No price yet | ||
| 18 | ?Zephyr 7B Beta | 1041.3±26.8 | 379 | 461 | 1102.1 | No price yet | ||
| 19 | Phi-3 Mini 4K InstructMicrosoft |
1043.7±14.9 | 378 | 1,706 | 1036.7 | No price yet | ||
| 20 | ?Nous Hermes 2 Mixtral 8x7B DPO | 1045.4±34.6 | 377 | 230 | 1101.4 | No price yet | ||
| 21 | ?Vicuna 13B | 1047±21.2 | 376 | 756 | 1057.9 | No price yet | ||
| 22 | Llama 3.2 1B InstructMeta |
1049.5±18.6 | 375 | 1,057 | 1032.5 | No price yet | ||
| 23 | WizardLM 13BMicrosoft |
1053.2±39.8 | 374 | 203 | 1090.2 | No price yet | ||
| 24 | ?OpenHermes 2.5 Mistral 7B | 1053.9±32.9 | 373 | 270 | 1096.9 | No price yet | ||
| 25 | Gemma 1.1 7B ITGoogle |
1055.3±12.5 | 372 | 2,649 | 1059.2 | No price yet | ||
| 26 | ?SmolLM2 1.7B Instruct | 1056±31.8 | 371 | 328 | 1031.2 | No price yet | ||
| 27 | Mistral 7B Instruct v0.2Mistral AI |
1059.9±15 | 370 | 1,592 | 1067.4 | No price yet | ||
| 28 | Llama 2 70B ChatMeta |
1061.7±12.6 | 369 | 2,813 | 1074.6 | No price yet | ||
| 29 | Llama 2 13B ChatMeta |
1063.9±17.2 | 368 | 1,193 | 1046.6 | No price yet | ||
| 30 |  GPT-3.5 Turbo (1106)OpenAI |
1069.4±20.8 | 367 | 933 | 1034.8 | No price yet | ||
| 31 |  Granite 3.0 2B InstructIBM |
1070±23.8 | 366 | 647 | 1046 | No price yet | ||
| 32 | ?Starling LM 7B Alpha | 1080.8±22.6 | 365 | 666 | 1098.9 | No price yet | ||
| 33 | ?OpenChat 3.5 (0106) | 1085.3±18.2 | 364 | 1,097 | 1092.2 | No price yet | ||
| 34 | Phi-3 Small 8K InstructMicrosoft |
1087.3±13.1 | 363 | 2,238 | 1082.6 | No price yet | ||
| 35 | Qwen1.5-7B-ChatAlibaba |
1089.8±29.8 | 362 | 372 | 1034.3 | No price yet | ||
| 36 |  DeepSeek LLM 67B ChatDeepSeek |
1091.7±36.1 | 361 | 243 | 1067.2 | No price yet | ||
| 37 | ?OpenChat 3.5 | 1092.6±28.1 | 360 | 414 | 1125.3 | No price yet | ||
| 38 | ?Yi 34B Chat | 1093.4±17.9 | 359 | 1,106 | 1107.8 | No price yet | ||
| 39 | WizardLM 70BMicrosoft |
1096.7±31.2 | 358 | 355 | 1148.9 | No price yet | ||
| 40 | Llama 3.2 3B InstructMeta |
1100.1±17.8 | 357 | 1,046 | 1093.9 | No price yet | ||
| 41 | Mixtral 8x7B Instruct v0.1Mistral AI |
1102.5±9.5 | 356 | 6,572 | 1108.5 | No price yet | ||
| 42 | ?Starling LM 7B Beta | 1103.8±15.3 | 355 | 1,676 | 1095.4 | No price yet | ||
| 43 | ?Zephyr ORPO 141B A35B v0.1 | 1104.5±27.8 | 354 | 448 | 1112.9 | No price yet | ||
| 44 | ?Tulu 2 DPO 70B | 1105.6±32.6 | 353 | 280 | 1104.7 | No price yet | ||
| 45 | ?DBRX Instruct Preview | 1111.3±12.4 | 352 | 3,489 | 1104 | No price yet | ||
| 46 | Qwen1.5-14B-ChatAlibaba |
1112.5±14.7 | 351 | 2,102 | 1090.3 | No price yet | ||
| 47 | Phi-3 Medium 4K InstructMicrosoft |
1119.8±12.9 | 350 | 2,150 | 1106.2 | No price yet | ||
| 48 | GPT-3.5 Turbo (0125)OpenAI |
1120.5±9.1 | 349 | 6,707 | 1092.1 | No price yet | ||
| 49 | Granite 3.0 8B InstructIBM |
1121.7±23.4 | 348 | 603 | 1070.4 | No price yet | ||
| 50 | Llama 3 8B InstructMeta |
1127.4±8.6 | 347 | 11,165 | 1149.6 | No price yet | ||
Show all 396 models (showing top 50)
Data: LMArena (lmarena-ai), CC-BY-4.0 · Longer prompts · 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).
Which LLM handles longer prompts best?
On LMArena's Longer prompts category (published Oct 22, 2026, 396 models), the top three are Gemini 4 Argon (High) (1537.8), Claude Opus 5.5 (High) (1532.8), Claude Fable 5.1 (Max) (1523.8).
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.