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: 31m ago

#1 Model

Claude Opus 4.6

Score: 100

Average Score

Across all ranked models

Models Ranked

106

With benchmark data

Weights:Arena Elo (100%)

Top Best for Roleplay Models by Weighted Score

Top 15 models by weighted score

LMMarketCap.com

#	Model	Provider	Score	Arena Elo
1	Claude Opus 4.6Anthropic	Anthropic	100	100
2	Gemini 3.1 Pro PreviewGoogle	Google	99	99
3	Muse Spark 1.1meta	meta	98.8	98.8
4	Claude Opus 4.7Anthropic	Anthropic	98.5	98.5
5	GPT-5.4OpenAI	OpenAI	97.5	97.5
6	GPT-5.2OpenAI	OpenAI	96.8	96.8
7	Claude Opus 4.8Anthropic	Anthropic	96.5	96.5
8	Gemini 3.5 FlashGoogle	Google	96.2	96.2
9	GPT-5.2 ChatOpenAI	OpenAI	96	96
10	GPT-5.5OpenAI	OpenAI	95.8	95.8
11	GLM 5.3 FlashZhipu AI	Zhipu AI	95.8	95.8
12	GPT-5.1OpenAI	OpenAI	95.8	95.8
13	Gemini 3 Flash PreviewGoogle	Google	95.7	95.7
14	Grok 4.5xAI	xAI	94.7	94.7
15	MiMo-V2.5-ProXiaomi	Xiaomi	94.5	94.5
16	GLM 5.1Zhipu AI	Zhipu AI	94.3	94.3
17	GPT-5OpenAI	OpenAI	94.2	94.2
18	Claude Sonnet 4.6Anthropic	Anthropic	93.3	93.3
19	Kimi K2.6Moonshot AI	Moonshot AI	93.3	93.3
20	Qwen3.6 Max PreviewAlibaba	Alibaba	93.3	93.3
21	GLM 5Zhipu AI	Zhipu AI	93	93
22	Gemini 3.5 Flash LiteGoogle	Google	92.7	92.7
23	Hy3Tencent	Tencent	92.7	92.7
24	Qwen3.7 PlusAlibaba	Alibaba	92.7	92.7
25	Claude Sonnet 4.5Anthropic	Anthropic	92	92
26	Gemma 4 31BGoogle	Google	91.8	91.8
27	Claude Opus 4.1Anthropic	Anthropic	91.7	91.7
28	Gemini 2.5 ProGoogle	Google	90.7	90.7
29	Qwen3.6 PlusAlibaba	Alibaba	90.5	90.5
30	Qwen3.5 397B A17BAlibaba	Alibaba	90.3	90.3
31	GLM 4.7Zhipu AI	Zhipu AI	90.3	90.3
32	MiniMax M3MiniMax	MiniMax	90.2	90.2
33	Inklingthinkingmachines	thinkingmachines	90	90
34	Gemma 4 26B A4B Google	Google	89.7	89.7
35	Qwen3.8 27BAlibaba	Alibaba	89.5	89.5
36	MiMo-V2.5Xiaomi	Xiaomi	89	89
37	GLM 5V TurboZhipu AI	Zhipu AI	88.8	88.8
38	Gemini 3.1 Flash Lite PreviewGoogle	Google	88.7	88.7
39	Claude Opus 4.5Anthropic	Anthropic	88.3	88.3
40	Mistral Medium 3.5Mistral AI	Mistral AI	87.7	87.7
41	DeepSeek V3.2DeepSeek	DeepSeek	87.5	87.5
42	GLM 4.6Zhipu AI	Zhipu AI	87.5	87.5
43	DeepSeek V3.2 ExpDeepSeek	DeepSeek	87	87
44	DeepSeek V3.1DeepSeek	DeepSeek	86.2	86.2
45	Qwen3.5-122B-A10BAlibaba	Alibaba	86.2	86.2
46	o3OpenAI	OpenAI	85.8	85.8
47	MiniMax M2.7MiniMax	MiniMax	85.8	85.8
48	DeepSeek V3.1 TerminusDeepSeek	DeepSeek	85.8	85.8
49	Qwen3 VL 235B A22B InstructAlibaba	Alibaba	85.7	85.7
50	Hy3 previewTencent	Tencent	85.5	85.5
51	GLM 4.5Zhipu AI	Zhipu AI	85.2	85.2
52	Qwen3.5-27BAlibaba	Alibaba	84.7	84.7
53	Inkling Smallthinkingmachines	thinkingmachines	84.2	84.2
54	Qwen3 Next 80B A3B InstructAlibaba	Alibaba	83.2	83.2
55	Qwen3.5-FlashAlibaba	Alibaba	82.8	82.8
56	Gemini 2.5 FlashGoogle	Google	82.5	82.5
57	Qwen3 VL 235B A22B ThinkingAlibaba	Alibaba	82.5	82.5
58	Qwen3.5-35B-A3BAlibaba	Alibaba	82.5	82.5
59	Step 3.5 FlashStepFun	StepFun	82.3	82.3
60	MiniMax M2.5MiniMax	MiniMax	81.7	81.7
61	GPT-5 MiniOpenAI	OpenAI	81.7	81.7
62	Claude Sonnet 4Anthropic	Anthropic	81.2	81.2
63	GPT-4.1 MiniOpenAI	OpenAI	80.5	80.5
64	R1 0528DeepSeek	DeepSeek	80	80
65	o4 MiniOpenAI	OpenAI	80	80
66	GLM 4.6VZhipu AI	Zhipu AI	79.7	79.7
67	GLM 4.5 AirZhipu AI	Zhipu AI	78.8	78.8
68	o3 MiniOpenAI	OpenAI	78.5	78.5
69	R1DeepSeek	DeepSeek	78.2	78.2
70	Qwen3 Next 80B A3B ThinkingAlibaba	Alibaba	78.2	78.2
71	Trinity Large Thinkingarcee-ai	arcee-ai	78.2	78.2
72	GLM 4.7 FlashZhipu AI	Zhipu AI	77.7	77.7
73	MiniMax M1MiniMax	MiniMax	77.3	77.3
74	o3 Mini HighOpenAI	OpenAI	77.2	77.2
75	GLM 4.5VZhipu AI	Zhipu AI	75.5	75.5
76	gpt-oss-120bOpenAI	OpenAI	75.3	75.3
77	o1OpenAI	OpenAI	75	75
78	Qwen3 8BAlibaba	Alibaba	74.5	74.5
79	Mercury 2Inception	Inception	74.3	74.3
80	MiniMax M2MiniMax	MiniMax	74.3	74.3
81	DeepSeek V3 0324DeepSeek	DeepSeek	74.2	74.2
82	GPT-5 NanoOpenAI	OpenAI	72.8	72.8
83	Nova 2 LiteAmazon	Amazon	72.7	72.7
84	Llama 4 MaverickMeta	Meta	70.8	70.8
85	GPT-4.1 NanoOpenAI	OpenAI	70.3	70.3
86	DeepSeek V3DeepSeek	DeepSeek	69.7	69.7
87	GPT-4o-mini (2024-07-18)OpenAI	OpenAI	69.7	69.7
88	gpt-oss-20bOpenAI	OpenAI	69.5	69.5
89	Mistral Large 2407Mistral AI	Mistral AI	69	69
90	GPT-4.1OpenAI	OpenAI	66.7	66.7
91	Granite 4.2 8BIBM	IBM	64.5	64.5
92	GPT-4oOpenAI	OpenAI	64.3	64.3
93	Mistral LargeMistral AI	Mistral AI	63.3	63.3
94	GPT-4OpenAI	OpenAI	62.7	62.7
95	Command ACohere	Cohere	60.2	60.2
96	Claude 3 HaikuAnthropic	Anthropic	60.2	60.2
97	Phi 4Microsoft	Microsoft	59.3	59.3
98	GPT-4 TurboOpenAI	OpenAI	59.2	59.2
99	Command R+ (08-2024)Cohere	Cohere	58.3	58.3
100	Llama 3.3 70B InstructMeta	Meta	57.2	57.2
101	Claude Haiku 4.5Anthropic	Anthropic	56.7	56.7
102	GPT-4o-miniOpenAI	OpenAI	53.7	53.7
103	Llama 3.1 8B InstructMeta	Meta	51.8	51.8
104	Llama 3.1 70B InstructMeta	Meta	49.7	49.7
105	Llama 3.2 3B InstructMeta	Meta	44.5	44.5
106	Llama 3.2 1B InstructMeta	Meta	35.2	35.2

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, Claude Opus 4.6 by Anthropic is currently the #1 ranked model for roleplay, with a weighted score of 100/100.

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

We currently rank 106 models that have relevant benchmark data for roleplay tasks.