Best AI Models for Roleplay

AI models ranked by roleplay and creative conversation quality using Arena Elo ratings, normalized to a 0-100 scale.

Last updated: just now

#1 Model

Gemini 2.5 Pro Preview 05-06

Score: 91

Average Score

67.3

Across all ranked models

Models Ranked

With benchmark data

Weights:Arena Elo (100%)

#	Model	Provider	Score	Arena Elo
1	Gemini 2.5 Pro Preview 05-06Google	Google	91	91
2	Gemini 2.5 ProGoogle	Google	90.7	90.7
3	Claude Opus 4.5Anthropic	Anthropic	88.3	88.3
4	o3OpenAI	OpenAI	88	88
5	Qwen3.5 397B A17BAlibaba	Alibaba	87.8	87.8
6	Grok 3xAI	xAI	87.2	87.2
7	Llama 4 MaverickMeta	Meta	86	86
8	Qwen3 235B A22B Thinking 2507Alibaba	Alibaba	85.5	85.5
9	Gemini 2.5 FlashGoogle	Google	85.2	85.2
10	GPT-4o (2024-11-20)OpenAI	OpenAI	80.2	80.2
11	GPT-4.1OpenAI	OpenAI	80	80
12	Qwen3 30B A3B Instruct 2507Alibaba	Alibaba	80	80
13	DeepSeek V3 0324DeepSeek	DeepSeek	79.3	79.3
14	o3 MiniOpenAI	OpenAI	78.5	78.5
15	Mistral Medium 3Mistral AI	Mistral AI	78.3	78.3
16	R1DeepSeek	DeepSeek	78.2	78.2
17	Claude Opus 4.6Anthropic	Anthropic	77.7	77.7
18	Qwen3 235B A22BAlibaba	Alibaba	77.5	77.5
19	Grok 3 Mini BetaxAI	xAI	76.7	76.7
20	MiniMax M1MiniMax	MiniMax	75.7	75.7
21	Claude 3.7 SonnetAnthropic	Anthropic	75.7	75.7
22	Gemini 2.0 FlashGoogle	Google	75.3	75.3
23	o1OpenAI	OpenAI	75	75
24	Qwen3 32BAlibaba	Alibaba	73.5	73.5
25	Llama 3.3 Nemotron Super 49B V1.5NVIDIA	NVIDIA	73.2	73.2
26	o3 Mini HighOpenAI	OpenAI	73	73
27	Claude Sonnet 4.6Anthropic	Anthropic	73	73
28	QwQ 32BAlibaba	Alibaba	71.8	71.8
29	Gemini 2.0 Flash LiteGoogle	Google	71.7	71.7
30	Qwen3 30B A3BAlibaba	Alibaba	70	70
31	DeepSeek V3DeepSeek	DeepSeek	69.7	69.7
32	GPT-4o (2024-05-13)OpenAI	OpenAI	67	67
33	GPT-4o-mini (2024-07-18)OpenAI	OpenAI	64.7	64.7
34	GPT-4oOpenAI	OpenAI	64.3	64.3
35	GPT-4o (2024-08-06)OpenAI	OpenAI	64.2	64.2
36	Llama 3.1 Nemotron 70B InstructNVIDIA	NVIDIA	64	64
37	Qwen2.5 72B InstructAlibaba	Alibaba	62	62
38	Claude 3.5 SonnetAnthropic	Anthropic	61.8	61.8
39	Mistral Large 2407Mistral AI	Mistral AI	61.5	61.5
40	Nova Pro 1.0Amazon	Amazon	60.2	60.2
41	GPT-4 TurboOpenAI	OpenAI	59.2	59.2
42	Mistral LargeMistral AI	Mistral AI	58.3	58.3
43	Llama 3.3 70B InstructMeta	Meta	57.2	57.2
44	Qwen2.5 Coder 32B InstructAlibaba	Alibaba	55.7	55.7
45	Command R+ (08-2024)Cohere	Cohere	55.5	55.5
46	Nova Lite 1.0Amazon	Amazon	55.5	55.5
47	Llama 3 70B InstructMeta	Meta	54	54
48	Phi 4Microsoft	Microsoft	53.7	53.7
49	GPT-4o-miniOpenAI	OpenAI	53.7	53.7
50	Nova Micro 1.0Amazon	Amazon	52.3	52.3
51	Claude 3 HaikuAnthropic	Anthropic	50	50
52	Llama 3.1 70B InstructMeta	Meta	49.7	49.7
53	Llama 3.1 8B InstructMeta	Meta	48.8	48.8
54	Claude 3.5 HaikuAnthropic	Anthropic	46.3	46.3
55	Llama 3 8B InstructMeta	Meta	45.2	45.2
56	Llama 3.2 3B InstructMeta	Meta	36.2	36.2
57	Llama 3.2 1B InstructMeta	Meta	27.8	27.8
58	Mistral 7B Instruct v0.1Mistral AI	Mistral AI	21.8	21.8

How scores are calculated

Each model's score is a weighted average of its available benchmark results. When a model is missing some benchmarks, the weights are re-normalized across the benchmarks that are available. All scores are on a 0-100 scale. Data sourced from official model cards, published papers, and third-party evaluation platforms.

Other Specialty Leaderboards

Best for Coding Best for Math Best for Reasoning Best for Writing Best for Instructions Best for Data Analysis Best for Multilingual

Frequently Asked Questions

Based on our benchmark analysis, Gemini 2.5 Pro Preview 05-06 by Google is currently the #1 ranked model for roleplay, with a weighted score of 91/100.

Models are ranked using a weighted average of Arena Elo benchmark scores. All scores are normalized to a 0-100 scale.

We currently rank 58 models that have relevant benchmark data for roleplay tasks.