/best/creative-writingupdated May 202686 models evaluated
Best AI for creative writing
Fiction, poetry, worldbuilding, and imaginative prose.
Creative writing is the hardest task to rank objectively. There is no ground truth. What we measure is a proxy: human preference on open-ended generation tasks, diversity of output across repeated prompts, and refusal rate on legitimate creative requests. The last metric matters — overly cautious models are frustrating for fiction writers.
What matters: voice range, structural coherence over long outputs, willingness to inhabit morally complex characters without breaking, and the ability to follow a creative brief without flattening it. The best creative models treat constraints as fuel, not as obstacles.
Our creative-writing rank weights writing quality (2.5×) and persona consistency (1.5×). Persona consistency is a proxy for character fidelity — a model that keeps losing track of who's speaking or drifts from established world rules produces unusable fiction.
Full rankings
All 86 models, scored for creative writing
| # | Model | Provider | Task score | $/in | $/out | Context |
|---|---|---|---|---|---|---|
| 01 | DeepSeek V4 Flash | DDeepSeek | 4.67/5.0 | $0.100 | $0.200 | 1.0M |
| 02 | MiMo-V2.5 | XXiaomi | 4.67/5.0 | $0.140 | $0.280 | 1.0M |
| 03 | GLM-4.7 Flash | ZZhipu AI | 4.67/5.0 | $0.060 | $0.400 | 203K |
| 04 | DeepSeek V4 Pro | DDeepSeek | 4.67/5.0 | $0.435 | $0.870 | 1.0M |
| 05 | Qwen: Qwen3.6 Plus | QQwen | 4.67/5.0 | $0.325 | $1.95 | 1M |
| 06 | R1 0528 | DDeepSeek | 4.67/5.0 | $0.500 | $2.15 | 164K |
| 07 | R1 | DDeepSeek | 4.67/5.0 | $0.700 | $2.50 | 164K |
| 08 | Grok 4.3 | XXAI | 4.67/5.0 | $1.25 | $2.50 | 1M |
| 09 | Gemini 3 Flash Preview | GGoogle | 4.67/5.0 | $0.500 | $3.00 | 1.0M |
| 10 | Qwen: Qwen3.6 27B | QQwen | 4.67/5.0 | $0.290 | $3.20 | 262K |
| 11 | MoonshotAI: Kimi K2.6 | MMoonshotAI | 4.67/5.0 | $0.730 | $3.49 | 262K |
| 12 | Qwen 3.7 Max | QQwen | 4.67/5.0 | $1.25 | $3.75 | 1M |
| 13 | Qwen: Qwen3.6 Max Preview | QQwen | 4.67/5.0 | $1.04 | $6.24 | 262K |
| 14 | Gemini 3.5 Flash | GGoogle | 4.67/5.0 | $1.50 | $9.00 | 1.0M |
| 15 | Gemini 3.1 Pro Preview | GGoogle | 4.67/5.0 | $2.00 | $12.00 | 1.0M |
| 16 | GPT-5.2 | OOpenAI | 4.67/5.0 | $1.75 | $14.00 | 400K |
| 17 | Claude Opus 4.7 | AAnthropic | 4.67/5.0 | $5.00 | $25.00 | 1M |
| 18 | GPT-5.5 | OOpenAI | 4.67/5.0 | $5.00 | $30.00 | 1.1M |
| 19 | Qwen: Qwen3 235B A22B Instruct 2507 | QQwen | 4.33/5.0 | $0.071 | $0.100 | 262K |
| 20 | Qwen: Qwen3.5-9B | QQwen | 4.33/5.0 | $0.040 | $0.150 | 262K |
| 21 | Ministral 3 8B 2512 | MMistral | 4.33/5.0 | $0.150 | $0.150 | 262K |
| 22 | Ministral 3 14B 2512 | MMistral | 4.33/5.0 | $0.200 | $0.200 | 262K |
| 23 | Gemma 4 31B | GGoogle | 4.33/5.0 | $0.120 | $0.370 | 262K |
| 24 | DeepSeek V3.2 | DDeepSeek | 4.33/5.0 | $0.252 | $0.378 | 131K |
| 25 | NVIDIA: Nemotron 3 Super | NNVIDIA | 4.33/5.0 | $0.090 | $0.450 | 1M |
| 26 | Grok 4.1 Fast | XXAI | 4.33/5.0 | $0.200 | $0.500 | 2M |
| 27 | DeepSeek V3.1 | DDeepSeek | 4.33/5.0 | $0.210 | $0.790 | 164K |
| 28 | MiMo-V2.5-Pro | XXiaomi | 4.33/5.0 | $0.435 | $0.870 | 1.0M |
| 29 | Qwen: Qwen3.6 35B A3B | QQwen | 4.33/5.0 | $0.140 | $1.00 | 262K |
| 30 | Qwen: Qwen3.6 Flash | QQwen | 4.33/5.0 | $0.188 | $1.13 | 1M |
| 31 | GPT-5.4 Nano | OOpenAI | 4.33/5.0 | $0.200 | $1.25 | 400K |
| 32 | Gemini 3.1 Flash Lite Preview | GGoogle | 4.33/5.0 | $0.250 | $1.50 | 1.0M |
| 33 | Google: Gemini 3.1 Flash Lite | GGoogle | 4.33/5.0 | $0.250 | $1.50 | 1.0M |
| 34 | Qwen: Qwen3.5 Plus 2026-04-20 | QQwen | 4.33/5.0 | $0.300 | $1.80 | 1M |
| 35 | GPT-5 Mini | OOpenAI | 4.33/5.0 | $0.250 | $2.00 | 400K |
| 36 | Devstral 2 2512 | MMistral | 4.33/5.0 | $0.400 | $2.00 | 262K |
| 37 | Mistral Medium 3.1 | MMistral | 4.33/5.0 | $0.400 | $2.00 | 131K |
| 38 | Xiaomi: MiMo-V2-Omni | XXiaomi | 4.33/5.0 | $0.400 | $2.00 | 262K |
| 39 | xAI: Grok Build 0.1 | XXAI | 4.33/5.0 | $1.00 | $2.00 | 256K |
| 40 | Gemini 2.5 Flash | GGoogle | 4.33/5.0 | $0.300 | $2.50 | 1.0M |
| 41 | Grok 4.20 | XXAI | 4.33/5.0 | $1.25 | $2.50 | 2M |
| 42 | GPT-5.4 Mini | OOpenAI | 4.33/5.0 | $0.750 | $4.50 | 400K |
| 43 | o4 Mini | OOpenAI | 4.33/5.0 | $1.10 | $4.40 | 200K |
| 44 | Mistral Medium 3.5 | MMistral | 4.33/5.0 | $1.50 | $7.50 | 262K |
| 45 | GPT-4.1 | OOpenAI | 4.33/5.0 | $2.00 | $8.00 | 1.0M |
| 46 | o3 | OOpenAI | 4.33/5.0 | $2.00 | $8.00 | 200K |
| 47 | GPT-5 | OOpenAI | 4.33/5.0 | $1.25 | $10.00 | 400K |
| 48 | Gemini 2.5 Pro | GGoogle | 4.33/5.0 | $1.25 | $10.00 | 1.0M |
| 49 | GPT-5.1 | OOpenAI | 4.33/5.0 | $1.25 | $10.00 | 400K |
| 50 | GPT-5.4 | OOpenAI | 4.33/5.0 | $2.50 | $15.00 | 1.1M |
| 51 | Claude Sonnet 4.6 | AAnthropic | 4.33/5.0 | $3.00 | $15.00 | 1M |
| 52 | Claude Opus 4.6 | AAnthropic | 4.33/5.0 | $5.00 | $25.00 | 1M |
| 53 | Anthropic: Claude Opus 4.8 | AAnthropic | 4.33/5.0 | $5.00 | $25.00 | 1M |
| 54 | Ministral 3 3B 2512 | MMistral | 4.00/5.0 | $0.100 | $0.100 | 131K |
| 55 | OpenAI: gpt-oss-120b | OOpenAI | 4.00/5.0 | $0.039 | $0.180 | 131K |
| 56 | Qwen: Qwen3 30B A3B Instruct 2507 | QQwen | 4.00/5.0 | $0.090 | $0.300 | 262K |
| 57 | Xiaomi: MiMo-V2-Flash | XXiaomi | 4.00/5.0 | $0.100 | $0.300 | 262K |
| 58 | Gemma 4 26B A4B | GGoogle | 4.00/5.0 | $0.060 | $0.330 | 262K |
| 59 | Gemini 2.5 Flash Lite | GGoogle | 4.00/5.0 | $0.100 | $0.400 | 1.0M |
| 60 | Grok 3 Mini | XXAI | 4.00/5.0 | $0.300 | $0.500 | 131K |
| 61 | Mistral Small 4 | MMistral | 4.00/5.0 | $0.150 | $0.600 | 262K |
| 62 | GPT-4.1 Mini | OOpenAI | 4.00/5.0 | $0.400 | $1.60 | 1.0M |
| 63 | GLM-4.7 | ZZhipu AI | 4.00/5.0 | $0.400 | $1.75 | 203K |
| 64 | Xiaomi: MiMo-V2-Pro | XXiaomi | 4.00/5.0 | $1.00 | $3.00 | 1.0M |
| 65 | Claude Haiku 4.5 | AAnthropic | 4.00/5.0 | $1.00 | $5.00 | 200K |
| 66 | Grok 4 | XXAI | 4.00/5.0 | $3.00 | $15.00 | 256K |
| 67 | Llama 4 Maverick | MMeta | 3.67/5.0 | $0.150 | $0.600 | 1.0M |
| 68 | DeepSeek V3.1 Terminus | DDeepSeek | 3.67/5.0 | $0.270 | $0.950 | 164K |
| 69 | GPT-4o | OOpenAI | 3.67/5.0 | $2.50 | $10.00 | 128K |
| 70 | Grok 3 | XXAI | 3.67/5.0 | $3.00 | $15.00 | 131K |
| 71 | OpenAI: gpt-oss-20b | OOpenAI | 3.33/5.0 | $0.030 | $0.140 | 131K |
| 72 | GPT-5 Nano | OOpenAI | 3.33/5.0 | $0.050 | $0.400 | 400K |
| 73 | GPT-4.1 Nano | OOpenAI | 3.33/5.0 | $0.100 | $0.400 | 1.0M |
| 74 | Grok Code Fast 1 | XXAI | 3.33/5.0 | $0.200 | $1.50 | 256K |
| 75 | NVIDIA: Nemotron 3 Nano 30B A3B | NNVIDIA | 3.00/5.0 | $0.050 | $0.200 | 262K |
| 76 | Mistral Small 3.2 24B | MMistral | 3.00/5.0 | $0.075 | $0.200 | 128K |
| 77 | Llama 4 Scout | MMeta | 3.00/5.0 | $0.080 | $0.300 | 10M |
| 78 | Llama 3.3 70B Instruct | MMeta | 3.00/5.0 | $0.100 | $0.320 | 131K |
| 79 | GPT-4o-mini | OOpenAI | 3.00/5.0 | $0.150 | $0.600 | 128K |
| 80 | Mistral Large 3 2512 | MMistral | 3.00/5.0 | $0.500 | $1.50 | 262K |
| 81 | Codestral 2508 | MMistral | 2.67/5.0 | $0.300 | $0.900 | 256K |
| 82 | Qwen: Qwen3.5-35B-A3B | QQwen | 2.67/5.0 | $0.139 | $1.00 | 262K |
| 83 | Devstral Medium | MMistral | 2.67/5.0 | $0.400 | $2.00 | 131K |
| 84 | Qwen: Qwen3 Coder 30B A3B Instruct | QQwen | 2.33/5.0 | $0.070 | $0.270 | 160K |
| 85 | Devstral Small 1.1 | MMistral | 2.33/5.0 | $0.100 | $0.300 | 131K |
| 86 | Mistral Small 3.1 24B | MMistral | 2.33/5.0 | $0.351 | $0.555 | 128K |
Pricing — top 5 for creative writing
D
4.67/5.0
X
4.67/5.0
Z
4.67/5.0
D
4.67/5.0
Q
4.67/5.0
The best AI for creative writing changes every month.
We'll email you when rankings shift, new models hit the top 5, or pricing cuts reshuffle the value leaders.
Get notified when models change
Price drops, new models, benchmark updates. One email per change, no spam.