Writing

Models ranked on creative writing, the writing apps and assistants built on them, and local setups for writing with open models on your own machine.

Models are ranked by LMArena Text Arena: Creative Writing (style control), then EQ-Bench Creative Writing v3 for models without a result on the first. Tools inherit the tier of the best model they offer. How we rank.

Best overallSClaude Opus 5.51,519 on LMArena Text Arena: Creative Writing (style control)
Best valueAGLM-5.3-FlashTier A for $0.24/M tokens
Best free startSClaudeFree tier · from $20/mo · via Claude Opus 5.5
Best local (≤24 GB GPU)BGemma 4 (31B)~19 GB VRAM · 1,419 on LMArena Text Arena: Creative Writing (style control)

Tier: how close to the best result on the task's main public benchmark
Outlined: a tool inherits the tier of the best model it offers
No public result yet
Cloud · Local: runs on your own hardware

Models

The engines themselves. Use them through an API or inside the apps on the right.

Cloud

Claude Opus 5.5Anthropic · $8/M tokens1,519 on LMArena Text Arena: Creative Writing (style control)$S

Gemini 4 ArgonGoogle · $4/M tokens1,516 on LMArena Text Arena: Creative Writing (style control)$S

Gemini 3.8 FlashGoogle · $1.5/M tokens1,487 on LMArena Text Arena: Creative Writing (style control)$S Muse WonderMeta · $8/M tokens1,485 on LMArena Text Arena: Creative Writing (style control)S

Claude Fable 5.1Anthropic · $20/M tokens1,485 on LMArena Text Arena: Creative Writing (style control)S

Gemini 3.1 Pro (Preview)Google · $4.5/M tokens1,481 on LMArena Text Arena: Creative Writing (style control)A Qwen3.8-MaxAlibaba (Qwen) · $3/M tokens1,469 on LMArena Text Arena: Creative Writing (style control)A

Claude Sonnet 5.5Anthropic · $4/M tokens1,465 on LMArena Text Arena: Creative Writing (style control)A Muse Spark 1.3Meta · $2/M tokens1,461 on LMArena Text Arena: Creative Writing (style control)A

GPT-6.1 SolOpenAI · $4/M tokens1,458 on LMArena Text Arena: Creative Writing (style control)A

GPT-6 AstraOpenAI · $20/M tokens1,450 on LMArena Text Arena: Creative Writing (style control)A

Gemini 3.5 Flash-LiteGoogle · $0.85/M tokens1,436 on LMArena Text Arena: Creative Writing (style control)$A

Qwen3.7-PlusAlibaba (Qwen) · $0.56/M tokens1,432 on LMArena Text Arena: Creative Writing (style control)$A

ERNIE 5.1Baidu1,427 on LMArena Text Arena: Creative Writing (style control)A

Grok 4.7xAI · $3/M tokens1,424 on LMArena Text Arena: Creative Writing (style control)B

GPT-6 LunaOpenAI · $0.2/M tokens1,410 on LMArena Text Arena: Creative Writing (style control)$B

GLM-5V-TurboZ.ai · $1.9/M tokens1,407 on LMArena Text Arena: Creative Writing (style control)B

Mistral Large 4 (preview)Mistral AI · $2.06/M tokens1,368 on LMArena Text Arena: Creative Writing (style control)C Aion 3.5AionLabs · $3.75/M tokens1,929 on EQ-Bench Creative Writing v3

Local

Kimi K3Moonshot AI · ~1600 GB VRAM1,461 on LMArena Text Arena: Creative Writing (style control)$A

GLM-5.3Z.ai · ~425 GB VRAM1,453 on LMArena Text Arena: Creative Writing (style control)$A

DeepSeek-V4-Pro-0813DeepSeek · ~920 GB VRAM1,446 on LMArena Text Arena: Creative Writing (style control)A

MiMo-V2.6-ProXiaomi · ~585 GB VRAM1,445 on LMArena Text Arena: Creative Writing (style control)A

DeepSeek-V4.1-Flash-1017DeepSeek · ~320 GB VRAM1,440 on LMArena Text Arena: Creative Writing (style control)$A

GLM-5.3-FlashZ.ai · ~185 GB VRAM1,434 on LMArena Text Arena: Creative Writing (style control)$B

Gemma 4 (31B)Google DeepMind · ~19 GB VRAM1,419 on LMArena Text Arena: Creative Writing (style control)$B

MiniMax-M3MiniMax · ~245 GB VRAM1,405 on LMArena Text Arena: Creative Writing (style control)B

Step 5 PreviewStepFun · ~346 GB VRAM1,403 on LMArena Text Arena: Creative Writing (style control)B

Mistral Medium 3.5Mistral AI · ~75 GB VRAM1,400 on LMArena Text Arena: Creative Writing (style control)B

MiMo-V2.6-FlashXiaomi · ~178 GB VRAM1,390 on LMArena Text Arena: Creative Writing (style control)B

InklingThinking Machines Lab · ~557 GB VRAM1,381 on LMArena Text Arena: Creative Writing (style control)B

Qwen3.8-27BAlibaba (Qwen) · ~17 GB VRAM1,366 on LMArena Text Arena: Creative Writing (style control)$B

Muse Glimmer 30BMeta · ~19 GB VRAM1,364 on LMArena Text Arena: Creative Writing (style control)C

Inkling SmallThinking Machines Lab · ~159 GB VRAM1,314 on LMArena Text Arena: Creative Writing (style control)C

Llama 4 MaverickMeta · ~230 GB VRAM1,307 on LMArena Text Arena: Creative Writing (style control)C

gpt-oss-120bOpenAI · ~80 GB VRAM1,279 on LMArena Text Arena: Creative Writing (style control)C

Hemmingway-1The Commons Lab (Altworld) · ~17 GB VRAM1,906 on EQ-Bench Creative Writing v3 C

gpt-oss-20bOpenAI · ~16 GB VRAM1,240 on LMArena Text Arena: Creative Writing (style control)$C

Qwen3.8-2.4T-A95BAlibaba (Qwen) · ~1370 GB VRAM1,843 on EQ-Bench Creative Writing v3

Apps

Ready-made apps with a UI: sign in or install, then start.

Cloud

ClaudeAnthropic · Free tier · $20/movia Claude Opus 5.5$S

GeminiGoogle · Free tier · $4.99/movia Gemini 4 Argon$A

KimiMoonshot AI · Free tiervia Kimi K3 A

Meta AIMeta · Free tiervia Muse Spark 1.3 A

ChatGPTOpenAI · Free tier · $20/movia GPT-6.1 Sol A Microsoft CopilotMicrosoft · Free tier · $9.99/movia GPT-6.1 Sol A

DeepSeekDeepSeek · Freevia DeepSeek-V4-Pro-0813$A

GrokSpaceXSI (formerly xAI) · Free tier · $8/movia Grok 4.7 B

Vibe (formerly Le Chat)Mistral AI · Free tier · $15/movia Mistral Medium 3.5 C

HemmingwayThe Commons Lab (Altworld)via Hemmingway-1 C

Qwen ChatAlibaba (Qwen team) · Freevia Qwen3.8-2.4T-A95B –

AI DungeonLatitudeNo public result yet –

DeepLDeepL · Free tierNo public result yet –

Gemini in Google WorkspaceGoogleNo public result yet –

GrammarlySuperhuman (formerly Grammarly Inc.) · Free tierNo public result yet –

Hemingway EditorHemingway · Free tierNo public result yet –

JasperJasper AI · $69/moNo public result yet –

Microsoft 365 CopilotMicrosoft · $25/moNo public result yet –

NotionNotion Labs · Free tier · $24/moNo public result yet –

NovelAIAnlatan · $10/moNo public result yet –

NovelcrafterNovelcrafter · $8/moNo public result yet – ProWritingAidProWritingAid · Free tierNo public result yet – SquiblerSquibler · Free tier · $30/moNo public result yet –

SudowriteSudowrite · $10/moNo public result yet –

TypeType.ai · Free tier · $8/moNo public result yet –

WordtuneAI21 Labs · Free tier · $4.89/moNo public result yet

Local

Open WebUIOpen WebUI Inc. · Freevia Kimi K3$A

TextGen (text-generation-webui)oobabooga · Freevia Kimi K3$A

JanMenlo Research · Freevia Kimi K3$A

SillyTavernSillyTavern (open-source project) · Freevia Kimi K3$A

KoboldCppLostRuins (open-source project) · Freevia Kimi K3$A

LM StudioElement Labs · Freevia Kimi K3$A

Msty StudioMsty · Freevia Kimi K3$–

LanguageToolLanguageTooler GmbH · Free tierNo public result yet –

mikupadlmg-anon · FreeNo public result yet

Runtimes & gateways

Run open models on your own machine, or reach many models through one API.

Cloud

OpenRouterOpenRouter · Free tiervia Claude Opus 5.5 –

DeepLDeepL · Free tierNo public result yet

Local

OllamaOllama · Freevia Kimi K3$A

llama.cppggml-org · Freevia Kimi K3$A

KoboldCppLostRuins (open-source project) · Freevia Kimi K3$

Quality vs. price

LMArena Text Arena: Creative Writing (style control) score against API price (USD per 1M tokens, blended 3:1 input:output, log scale). Up and to the left is better; the line joins the models that nothing else beats on both.

Closed models
Open-weights models
Best-value frontier

Show as table

Model	Weights	LMArena Text Arena: Creative Writing (style control)	Price	Frontier
Claude Opus 5.5	Closed	1,519	$8/M tokens	Yes
Gemini 4 Argon	Closed	1,516	$4/M tokens	Yes
Gemini 3.8 Flash	Closed	1,487	$1.5/M tokens	Yes
Muse Wonder	Closed	1,485	$8/M tokens
Claude Fable 5.1	Closed	1,485	$20/M tokens
Gemini 3.1 Pro (Preview)	Closed	1,481	$4.5/M tokens
Qwen3.8-Max	Closed	1,469	$3/M tokens
Claude Sonnet 5.5	Closed	1,465	$4/M tokens
Kimi K3	Open	1,461	$6/M tokens
Muse Spark 1.3	Closed	1,461	$2/M tokens
GPT-6.1 Sol	Closed	1,458	$4/M tokens
GLM-5.3	Open	1,453	$2.15/M tokens
GPT-6 Astra	Closed	1,450	$20/M tokens
DeepSeek-V4-Pro-0813	Open	1,446	$1.98/M tokens
MiMo-V2.6-Pro	Open	1,445	$0.54/M tokens	Yes
DeepSeek-V4.1-Flash-1017	Open	1,440	$0.52/M tokens	Yes
Gemini 3.5 Flash-Lite	Closed	1,436	$0.85/M tokens
GLM-5.3-Flash	Open	1,434	$0.24/M tokens	Yes
Qwen3.7-Plus	Closed	1,432	$0.56/M tokens
Grok 4.7	Closed	1,424	$3/M tokens
Gemma 4 (31B)	Open	1,419	$0.21/M tokens	Yes
GPT-6 Luna	Closed	1,410	$0.2/M tokens	Yes
GLM-5V-Turbo	Closed	1,407	$1.9/M tokens
MiniMax-M3	Open	1,405	$0.52/M tokens
Step 5 Preview	Open	1,403	$1.43/M tokens
Mistral Medium 3.5	Open	1,400	$3/M tokens
MiMo-V2.6-Flash	Open	1,390	$0.18/M tokens	Yes
Inkling	Open	1,381	$1.76/M tokens
Mistral Large 4 (preview)	Closed	1,368	$2.06/M tokens
Qwen3.8-27B	Open	1,366	$0.96/M tokens
Muse Glimmer 30B	Open	1,364	$0.52/M tokens
Inkling Small	Open	1,314	$0.64/M tokens
Llama 4 Maverick	Open	1,307	$0.55/M tokens
gpt-oss-120b	Open	1,279	$0.26/M tokens
gpt-oss-20b	Open	1,240	$0.13/M tokens	Yes

Everything, in order

All writing options in quality order
Tier	Name	Approach	Runs	Rated by	Price	Skill
S	Claude Opus 5.5 Anthropic	Models	Cloud	1,519 on LMArena Text Arena: Creative Writing (style control)	$8/M tokens	—
S	Gemini 4 Argon Google	Models	Cloud	1,516 on LMArena Text Arena: Creative Writing (style control)	$4/M tokens	—
S	Gemini 3.8 Flash Google	Models	Cloud	1,487 on LMArena Text Arena: Creative Writing (style control)	$1.5/M tokens	—
S	Muse Wonder Meta	Models	Cloud	1,485 on LMArena Text Arena: Creative Writing (style control)	$8/M tokens	—
S	Claude Fable 5.1 Anthropic	Models	Cloud	1,485 on LMArena Text Arena: Creative Writing (style control)	$20/M tokens	—
S	Gemini 3.1 Pro (Preview) Google	Models	Cloud	1,481 on LMArena Text Arena: Creative Writing (style control)	$4.5/M tokens	—
S	Claude Anthropic	Apps	Cloud	via Claude Opus 5.5	Free tier · $20/mo	beginner
S	OpenRouter OpenRouter	Runtimes & gateways	Cloud	via Claude Opus 5.5	Free tier	intermediate
S	Gemini Google	Apps	Cloud	via Gemini 4 Argon	Free tier · $4.99/mo	beginner
A	Qwen3.8-Max Alibaba (Qwen)	Models	Cloud	1,469 on LMArena Text Arena: Creative Writing (style control)	$3/M tokens	—
A	Claude Sonnet 5.5 Anthropic	Models	Cloud	1,465 on LMArena Text Arena: Creative Writing (style control)	$4/M tokens	—
A	Kimi K3 Moonshot AI	Models	Local	1,461 on LMArena Text Arena: Creative Writing (style control)	~1600 GB VRAM	—
A	Muse Spark 1.3 Meta	Models	Cloud	1,461 on LMArena Text Arena: Creative Writing (style control)	$2/M tokens	—
A	GPT-6.1 Sol OpenAI	Models	Cloud	1,458 on LMArena Text Arena: Creative Writing (style control)	$4/M tokens	—
A	GLM-5.3 Z.ai	Models	Local	1,453 on LMArena Text Arena: Creative Writing (style control)	~425 GB VRAM	—
A	GPT-6 Astra OpenAI	Models	Cloud	1,450 on LMArena Text Arena: Creative Writing (style control)	$20/M tokens	—
A	DeepSeek-V4-Pro-0813 DeepSeek	Models	Local	1,446 on LMArena Text Arena: Creative Writing (style control)	~920 GB VRAM	—
A	MiMo-V2.6-Pro Xiaomi	Models	Local	1,445 on LMArena Text Arena: Creative Writing (style control)	~585 GB VRAM	—
A	DeepSeek-V4.1-Flash-1017 DeepSeek	Models	Local	1,440 on LMArena Text Arena: Creative Writing (style control)	~320 GB VRAM	—
A	Gemini 3.5 Flash-Lite Google	Models	Cloud	1,436 on LMArena Text Arena: Creative Writing (style control)	$0.85/M tokens	—
A	GLM-5.3-Flash Z.ai	Models	Local	1,434 on LMArena Text Arena: Creative Writing (style control)	~185 GB VRAM	—
A	Qwen3.7-Plus Alibaba (Qwen)	Models	Cloud	1,432 on LMArena Text Arena: Creative Writing (style control)	$0.56/M tokens	—
A	ERNIE 5.1 Baidu	Models	Cloud	1,427 on LMArena Text Arena: Creative Writing (style control)	—	—
A	Grok 4.7 xAI	Models	Cloud	1,424 on LMArena Text Arena: Creative Writing (style control)	$3/M tokens	—
A	Ollama Ollama	Runtimes & gateways	Local	via Kimi K3	Free	intermediate
A	Open WebUI Open WebUI Inc.	Apps	Local	via Kimi K3	Free	intermediate
A	llama.cpp ggml-org	Runtimes & gateways	Local	via Kimi K3	Free	advanced
A	TextGen (text-generation-webui) oobabooga	Apps	Local	via Kimi K3	Free	intermediate
A	Jan Menlo Research	Apps	Local	via Kimi K3	Free	beginner
A	SillyTavern SillyTavern (open-source project)	Apps	Local	via Kimi K3	Free	advanced
A	KoboldCpp LostRuins (open-source project)	Apps + Runtimes & gateways	Local	via Kimi K3	Free	intermediate
A	Kimi Moonshot AI	Apps	Cloud	via Kimi K3	Free tier	beginner
A	LM Studio Element Labs	Apps	Local	via Kimi K3	Free	beginner
A	Meta AI Meta	Apps	Cloud	via Muse Spark 1.3	Free tier	beginner
A	Msty Studio Msty	Apps	Local	via Kimi K3	Free	beginner
A	ChatGPT OpenAI	Apps	Cloud	via GPT-6.1 Sol	Free tier · $20/mo	beginner
A	Microsoft Copilot Microsoft	Apps	Cloud	via GPT-6.1 Sol	Free tier · $9.99/mo	beginner
A	DeepSeek DeepSeek	Apps	Cloud	via DeepSeek-V4-Pro-0813	Free	beginner
A	Grok SpaceXSI (formerly xAI)	Apps	Cloud	via Grok 4.7	Free tier · $8/mo	beginner
B	Gemma 4 (31B) Google DeepMind	Models	Local	1,419 on LMArena Text Arena: Creative Writing (style control)	~19 GB VRAM	—
B	GPT-6 Luna OpenAI	Models	Cloud	1,410 on LMArena Text Arena: Creative Writing (style control)	$0.2/M tokens	—
B	GLM-5V-Turbo Z.ai	Models	Cloud	1,407 on LMArena Text Arena: Creative Writing (style control)	$1.9/M tokens	—
B	MiniMax-M3 MiniMax	Models	Local	1,405 on LMArena Text Arena: Creative Writing (style control)	~245 GB VRAM	—
B	Step 5 Preview StepFun	Models	Local	1,403 on LMArena Text Arena: Creative Writing (style control)	~346 GB VRAM	—
B	Mistral Medium 3.5 Mistral AI	Models	Local	1,400 on LMArena Text Arena: Creative Writing (style control)	~75 GB VRAM	—
B	MiMo-V2.6-Flash Xiaomi	Models	Local	1,390 on LMArena Text Arena: Creative Writing (style control)	~178 GB VRAM	—
B	Inkling Thinking Machines Lab	Models	Local	1,381 on LMArena Text Arena: Creative Writing (style control)	~557 GB VRAM	—
B	Mistral Large 4 (preview) Mistral AI	Models	Cloud	1,368 on LMArena Text Arena: Creative Writing (style control)	$2.06/M tokens	—
B	Qwen3.8-27B Alibaba (Qwen)	Models	Local	1,366 on LMArena Text Arena: Creative Writing (style control)	~17 GB VRAM	—
B	Muse Glimmer 30B Meta	Models	Local	1,364 on LMArena Text Arena: Creative Writing (style control)	~19 GB VRAM	—
B	Vibe (formerly Le Chat) Mistral AI	Apps	Cloud	via Mistral Medium 3.5	Free tier · $15/mo	beginner
C	Inkling Small Thinking Machines Lab	Models	Local	1,314 on LMArena Text Arena: Creative Writing (style control)	~159 GB VRAM	—
C	Llama 4 Maverick Meta	Models	Local	1,307 on LMArena Text Arena: Creative Writing (style control)	~230 GB VRAM	—
C	gpt-oss-120b OpenAI	Models	Local	1,279 on LMArena Text Arena: Creative Writing (style control)	~80 GB VRAM	—
C	Aion 3.5 AionLabs	Models	Cloud	1,929 on EQ-Bench Creative Writing v3	$3.75/M tokens	—
C	Hemmingway-1 The Commons Lab (Altworld)	Models	Local	1,906 on EQ-Bench Creative Writing v3	~17 GB VRAM	—
C	gpt-oss-20b OpenAI	Models	Local	1,240 on LMArena Text Arena: Creative Writing (style control)	~16 GB VRAM	—
C	Qwen3.8-2.4T-A95B Alibaba (Qwen)	Models	Local	1,843 on EQ-Bench Creative Writing v3	~1370 GB VRAM	—
C	Hemmingway The Commons Lab (Altworld)	Apps	Cloud	via Hemmingway-1	—	beginner
C	Qwen Chat Alibaba (Qwen team)	Apps	Cloud	via Qwen3.8-2.4T-A95B	Free	beginner
–	LanguageTool LanguageTooler GmbH	Apps	Local	No public result yet	Free tier	beginner
–	mikupad lmg-anon	Apps	Local	No public result yet	Free	advanced
–	AI Dungeon Latitude	Apps	Cloud	No public result yet	—	beginner
–	DeepL DeepL	Apps + Runtimes & gateways	Cloud	No public result yet	Free tier	beginner
–	Gemini in Google Workspace Google	Apps	Cloud	No public result yet	—	beginner
–	Grammarly Superhuman (formerly Grammarly Inc.)	Apps	Cloud	No public result yet	Free tier	beginner
–	Hemingway Editor Hemingway	Apps	Cloud	No public result yet	Free tier	beginner
–	Jasper Jasper AI	Apps	Cloud	No public result yet	$69/mo	beginner
–	Microsoft 365 Copilot Microsoft	Apps	Cloud	No public result yet	$25/mo	beginner
–	Notion Notion Labs	Apps	Cloud	No public result yet	Free tier · $24/mo	beginner
–	NovelAI Anlatan	Apps	Cloud	No public result yet	$10/mo	beginner
–	Novelcrafter Novelcrafter	Apps	Cloud	No public result yet	$8/mo	intermediate
–	ProWritingAid ProWritingAid	Apps	Cloud	No public result yet	Free tier	beginner
–	Squibler Squibler	Apps	Cloud	No public result yet	Free tier · $30/mo	beginner
–	Sudowrite Sudowrite	Apps	Cloud	No public result yet	$10/mo	beginner
–	Type Type.ai	Apps	Cloud	No public result yet	Free tier · $8/mo	beginner
–	Wordtune AI21 Labs	Apps	Cloud	No public result yet	Free tier · $4.89/mo	beginner

Benchmarks used

LMArena Text Arena: Creative Writing (style control) by LMArena (arena.ai) (blind human votes, Arena score (Bradley-Terry rating from pairwise human votes, creative-writing prompts only, style-controlled view)) · snapshot 15 Oct 2026
Read from LMArena's official Hugging Face dataset (text_style_control, latest split, category creative_writing). 414 entries; covers almost every current family incl. open ones (GLM, Kimi, DeepSeek, MiMo, MiniMax, Gemma, Qwen3.8-27B, Muse Glimmer, gpt-oss). Not listed: Claude Haiku 5.5, Muse Spark Open, the open Qwen3.8-2.4T-A95B checkpoint (the API 'qwen3.8-max' is). Several top entries have under 2k votes (e.g. #1 Claude Opus 5.5 ~1.9k).
EQ-Bench Creative Writing v3 by EQ-Bench (Sam Paech) (automated tests, Elo score from LLM-judged pairwise comparisons of creative-writing outputs (rubric score shown alongside)) · snapshot 24 Sep 2026
Creative Writing v3 (v2 is legacy; no v4 as of 2026-10-19). Data read from the CSV embedded in https://eqbench.com/creative_writing.js (Last-Modified 2026-09-24, unchanged on 2026-10-19); the page shows no update date. 140 entries; covers most current families but lists only previous versions of GPT Sol (gpt-6-sol), Claude Sonnet (claude-sonnet-5) and GPT Luna (gpt-5.6-luna) and lacks GPT-6.1 Sol, GPT-6 Luna, Claude Sonnet 5.5, Gemini 4 Argon and Claude Haiku 5.5; includes stealth/anonymous entries (e.g. ox-alpha; not mapped). LLM-judged (automated), not human preference; Elo spread is far wider than on human arenas (#1 2173 vs last 200). The same author runs a separate Longform Creative Writing board (creative_writing_longform.html), updated the same day.

Writing

Quality vs. price

Quality vs. setup

Everything, in order

Benchmarks used