Chatbot Arena

This leaderboard is based on the following benchmarks.

• Chatbot Arena - a crowdsourced, randomized battle platform for large language models (LLMs). We use 2.7M+ user votes to compute Elo ratings.
• MMLU - a test to measure a model’s multitask accuracy on 57 tasks.
• Arena-Hard-Auto - an automatic evaluation tool for instruction-tuned LLMs.

| Vote | Blog | GitHub | Paper | Dataset | Twitter | Discord |

Best Open LM

Model	Arena Elo	MMLU	License
DeepSeek-R1	1362	90.8	MIT
DeepSeek-V3	1318	88.5	DeepSeek
Qwen2.5-72B-Instruct	1257	86.8	Qwen
Llama-3.3-70B-Instruct	1256	86	Llama 3.3

Full Leaderboard

Model	Arena Elo	Coding	Vision	Arena Hard	MMLU	Votes	Organization	License
🥇🏆chocolate (Early Grok-3)	1403	1403				9992	xAI	Proprietary
🥈 Gemini-2.0-Flash-Thinking-Exp-01-21	1385	1368	1282			15083	Google	Proprietary
🥈 Gemini-2.0-Pro-Exp-02-05	1380	1373	1251			13000	Google	Proprietary
🥈 ChatGPT-4o-latest (2025-01-29)	1377	1364	1276			13470	OpenAI	Proprietary
🥈 DeepSeek-R1	1362	1369				6581	DeepSeek	MIT
🥈 Gemini-2.0-Flash-001	1358	1355	1232			10862	Google	Proprietary
🥈 o1-2024-12-17	1352	1360	1216	90.4		17248	OpenAI	Proprietary
🥈 Qwen2.5-Max	1334	1337				9282	Alibaba	Proprietary
🥈 o3-mini-high	1332	1373				5954	OpenAI	Proprietary
🥉 DeepSeek-V3	1318	1316				19461	DeepSeek	DeepSeek
🥉 Qwen-Plus-0125	1311	1317				5112	Alibaba	Proprietary
🥉 GLM-4-Plus-0111	1310	1286				5134	Zhipu	Proprietary
🥉 Gemini-2.0-Flash-Lite-Preview-02-05	1309	1323	1154			10262	Google	Proprietary
🥉 o3-mini	1306	1355				12179	OpenAI	Proprietary
🥉 o1-mini	1304	1353		92		54944	OpenAI	Proprietary
🥉 Step-2-16K-Exp	1304	1295				5130	StepFun	Proprietary
🥉 Gemini-1.5-Pro-002	1302	1291	1220			54970	Google	Proprietary
🥉 Grok-2-08-13	1288	1282				67045	xAI	Proprietary
🥉 Yi-Lightning	1287	1302		81.5		28958	01 AI	Proprietary
🥉 Claude 3.5 Sonnet (20241022)	1283	1326	1184	85.2	88.7	56346	Anthropic	Proprietary
Deepseek-v2.5-1210	1279	1296				7245	DeepSeek	DeepSeek
Athene-v2-Chat-72B	1275	1300		85		26092	NexusFlow	NexusFlow
GPT-4o-mini-2024-07-18	1273	1283	1124	74.94	82	65335	OpenAI	Proprietary
Gemini-1.5-Flash-002	1271	1254	1205			36993	Google	Proprietary
Llama-3.1-405B-Instruct-bf16	1269	1280			88.6	31499	Meta	Llama 3.1
Llama-3.1-Nemotron-70B-Instruct	1268	1271		84.9		7601	Nvidia	Llama 3.1
Grok-2-Mini-08-13	1266	1262				55424	xAI	Proprietary
Yi-Lightning-lite	1264	1267				17059	01 AI	Proprietary
Qwen2.5-72B-Instruct	1257	1283		78		41543	Alibaba	Qwen
GPT-4-Turbo-2024-04-09	1256	1263	1151	82.63		102119	OpenAI	Proprietary
Llama-3.3-70B-Instruct	1256	1258				25878	Meta	Llama-3.3
Mistral-Large-2407	1251	1269		70.42		48201	Mistral	Mistral Research
GPT-4-1106-preview	1250	1253				103725	OpenAI	Proprietary
Athene-70B	1250	1253		77.6		20604	NexusFlow	CC-BY-NC-4.0
Llama-3.1-70B-Instruct	1248	1251		55.73	86	58750	Meta	Llama 3.1
Mistral-Large-2411	1247	1266				20816	Mistral	MRL
Claude 3 Opus	1247	1250	1076	60.36	86.8	202650	Anthropic	Proprietary
Amazon Nova Pro 1.0	1246	1259	1043			18006	Amazon	Proprietary
GPT-4-0125-preview	1245	1244		77.96		97041	OpenAI	Proprietary
Llama-3.1-Tulu-3-70B	1244	1232				3014	Ai2	Llama 3.1
Yi-Large-preview	1240	1245		71.48		51652	01 AI	Proprietary
Claude 3.5 Haiku (20241022)	1236	1263				21000	Anthropic	Propretary
Reka-Core-20240904	1235	1222				7935	Reka AI	Proprietary
Reka-Core-20240722	1231	1208				13286	Reka AI	Proprietary
Qwen-Plus-0828	1227	1245				14612	Alibaba	Proprietary
Gemini-1.5-Flash-001	1227	1232	1072	49.61	78.9	65657	Google	Proprietary
Jamba-1.5-Large	1221	1227			81.2	9120	AI21 Labs	Jamba Open
Deepseek-v2-API-0628	1220	1242				19499	DeepSeek AI	DeepSeek
Gemma-2-27B-it	1220	1210		57.51		77543	Google	Gemma license
Qwen2.5-Coder-32B-Instruct	1217	1261				5724	Alibaba	Apache 2.0
Amazon Nova Lite 1.0	1216	1234	1060			15984	Amazon	Proprietary
Gemma-2-9B-it-SimPO	1216	1196				10549	Princeton	MIT
Command R+ (08-2024)	1215	1181				10539	Cohere	CC-BY-NC-4.0
Deepseek-Coder-v2-0724	1214	1266		62.3		11723	DeepSeek	Proprietary
Gemini-1.5-Flash-8B-001	1213	1208	1106			37680	Google	Proprietary
Yi-Large	1212	1220		63.7		16629	01 AI	Proprietary
Llama-3.1-Nemotron-51B-Instruct	1211	1211				3891	Nvidia	Llama 3.1
Mistral-Small-24B-Instruct-2501	1210	1215				6280	Mistral	Apache 2.0
Nemotron-4-340B-Instruct	1209	1198				20608	Nvidia	NVIDIA Open Model
Aya-Expanse-32B	1209	1193				28750	Cohere	CC-BY-NC-4.0
Gemini App (2024-01-24)	1208	1171				11827	Google	Proprietary
GLM-4-0520	1207	1216		63.84		10215	Zhipu AI	Proprietary
Llama-3-70B-Instruct	1206	1200		46.57	82	163745	Meta	Llama 3
Reka-Flash-20240904	1205	1191				8127	Reka AI	Proprietary
Gemini-1.5-Flash-8B-Exp-0827	1205	1189	1112			25350	Google	Proprietary
Phi-4	1204	1226				13156	Microsoft	MIT
Claude 3 Sonnet	1201	1213	1048	46.8	79	113016	Anthropic	Proprietary
Reka-Flash-20240722	1201	1187				13722	Reka AI	Proprietary
Reka-Core-20240501	1200	1190	1015		83.2	62563	Reka AI	Proprietary
Amazon Nova Micro 1.0	1198	1210				16043	Amazon	Proprietary
Gemma-2-9B-it	1191	1173				55285	Google	Gemma license
Command R+ (04-2024)	1190	1164		33.07		80868	Cohere	CC-BY-NC-4.0
Hunyuan-Standard-256K	1189	1227				2898	Tencent	Proprietary
Qwen2-72B-Instruct	1187	1187		46.86	84.2	38875	Alibaba	Qianwen LICENSE
GPT-4-0314	1186	1195		50	86.4	55951	OpenAI	Proprietary
Llama-3.1-Tulu-3-8B	1185	1178				3073	Ai2	Llama 3.1
GLM-4-0116	1183	1191		55.72		7580	Zhipu AI	Proprietary
Qwen-Max-0428	1183	1189				25676	Alibaba	Proprietary
Ministral-8B-2410	1182	1201				5108	Mistral	MRL
Aya-Expanse-8B	1180	1165				10404	Cohere	CC-BY-NC-4.0
Command R (08-2024)	1180	1162				10845	Cohere	CC-BY-NC-4.0
Claude 3 Haiku	1179	1189	1000	41.47	75.2	122289	Anthropic	Proprietary
DeepSeek-Coder-V2-Instruct	1178	1239				15749	DeepSeek AI	DeepSeek License
Llama-3.1-8B-Instruct	1176	1186		21.34	73	52633	Meta	Llama 3.1
Jamba-1.5-Mini	1176	1181			69.7	9271	AI21 Labs	Jamba Open
Reka-Flash-Preview-20240611	1165	1155	1024			20411	Reka AI	Proprietary
GPT-4-0613	1163	1167		37.9		91608	OpenAI	Proprietary
Qwen1.5-110B-Chat	1161	1175			80.4	27453	Alibaba	Qianwen LICENSE
Mistral-Large-2402	1157	1170		37.71	81.2	64903	Mistral	Proprietary
Yi-1.5-34B-Chat	1157	1162			76.8	25121	01 AI	Apache-2.0
Reka-Flash-21B-online	1156	1147				16016	Reka AI	Proprietary
QwQ-32B-Preview	1153	1146				3413	Alibaba	Apache 2.0
Llama-3-8B-Instruct	1152	1146		20.56	68.4	109218	Meta	Llama 3
InternLM2.5-20B-chat	1149	1158				10595	InternLM	Other
Claude-1	1149	1136			77	21158	Anthropic	Proprietary
Command R (04-2024)	1149	1123		17.02		56356	Cohere	CC-BY-NC-4.0
Mixtral-8x22b-Instruct-v0.1	1148	1152		36.36	77.8	53773	Mistral	Apache 2.0
Mistral Medium	1148	1152		31.9	75.3	35552	Mistral	Proprietary
Reka-Flash-21B	1148	1141			73.5	25807	Reka AI	Proprietary
Qwen1.5-72B-Chat	1147	1160		36.12	77.5	40638	Alibaba	Qianwen LICENSE
Gemma-2-2b-it	1144	1108			51.3	47032	Google	Gemma license
Granite-3.1-8B-Instruct	1143	1171				3302	IBM	Apache 2.0
Claude-2.0	1132	1135		23.99	78.5	12758	Anthropic	Proprietary
Gemini-1.0-Pro-001	1131	1103			71.8	18789	Google	Proprietary
Zephyr-ORPO-141b-A35b-v0.1	1127	1124				4859	HuggingFace	Apache 2.0
Qwen1.5-32B-Chat	1125	1149			73.4	22766	Alibaba	Qianwen LICENSE
Mistral-Next	1124	1132		27.37		12372	Mistral	Proprietary
Phi-3-Medium-4k-Instruct	1123	1125		33.37	78	26092	Microsoft	MIT
Granite-3.1-2B-Instruct	1120	1147				3385	IBM	Apache 2.0
Starling-LM-7B-beta	1119	1129		23.01		16668	Nexusflow	Apache-2.0
Claude-2.1	1118	1132		22.77		37688	Anthropic	Proprietary
GPT-3.5-Turbo-0613	1117	1135		24.82		38940	OpenAI	Proprietary
Mixtral-8x7B-Instruct-v0.1	1114	1114		23.4	70.6	76103	Mistral	Apache 2.0
Claude-Instant-1	1111	1109			73.4	20618	Anthropic	Proprietary
Yi-34B-Chat	1111	1106		23.15	73.5	15924	01 AI	Yi License
Gemini Pro	1110	1091		17.8	71.8	6559	Google	Proprietary
Qwen1.5-14B-Chat	1109	1126			67.6	18667	Alibaba	Qianwen LICENSE
GPT-3.5-Turbo-0125	1106	1124		23.34		68856	OpenAI	Proprietary
GPT-3.5-Turbo-0314	1106	1115		18.05	70	5648	OpenAI	Proprietary
WizardLM-70B-v1.0	1106	1071			63.7	8380	Microsoft	Llama 2
DBRX-Instruct-Preview	1103	1118		24.63	73.7	33720	Databricks	DBRX LICENSE
Llama-3.2-3B-Instruct	1103	1080				8405	Meta	Llama 3.2
Phi-3-Small-8k-Instruct	1102	1107		29.77	75.7	18479	Microsoft	MIT
Tulu-2-DPO-70B	1099	1093		14.99		6656	AllenAI/UW	AI2 ImpACT Low-risk
Granite-3.0-8B-Instruct	1093	1097				7003	IBM	Apache 2.0
Llama-2-70B-chat	1093	1072		11.55	63	39614	Meta	Llama 2
OpenChat-3.5-0106	1091	1102			65.8	12981	OpenChat	Apache-2.0
Vicuna-33B	1091	1067		8.63	59.2	22944	LMSYS	Non-commercial
Snowflake Arctic Instruct	1090	1077			67.3	34176	Snowflake	Apache 2.0
Starling-LM-7B-alpha	1088	1080		12.8	63.9	10415	UC Berkeley	CC-BY-NC-4.0
Gemma-1.1-7B-it	1084	1084			64.3	25071	Google	Gemma license
Nous-Hermes-2-Mixtral-8x7B-DPO	1084	1079				3834	NousResearch	Apache-2.0
NV-Llama2-70B-SteerLM-Chat	1081	1023			68.5	3637	Nvidia	Llama 2
pplx-70B-online	1078	1028				6896	Perplexity AI	Proprietary
DeepSeek-LLM-67B-Chat	1077	1079			71.3	4984	DeepSeek AI	DeepSeek License
OpenChat-3.5	1076	1054			64.3	8106	OpenChat	Apache-2.0
Granite-3.0-2B-Instruct	1074	1088				7185	IBM	Apache 2.0
OpenHermes-2.5-Mistral-7B	1074	1058				5088	NousResearch	Apache-2.0
Mistral-7B-Instruct-v0.2	1072	1074				20050	Mistral	Apache-2.0
Qwen1.5-7B-Chat	1070	1089			61	4863	Alibaba	Qianwen LICENSE
Phi-3-Mini-4K-Instruct-June-24	1070	1082			70.9	12807	Microsoft	MIT
GPT-3.5-Turbo-1106	1068	1095				17031	OpenAI	Proprietary
Phi-3-Mini-4k-Instruct	1066	1086			68.8	21094	Microsoft	MIT
Llama-2-13b-chat	1063	1051			53.6	19730	Meta	Llama 2
Dolphin-2.2.1-Mistral-7B	1063	1025				1714	Cognitive Computations	Apache-2.0
SOLAR-10.7B-Instruct-v1.0	1062	1047			66.2	4288	Upstage AI	CC-BY-NC-4.0
WizardLM-13b-v1.2	1058	1026			52.7	7174	Microsoft	Llama 2
Llama-3.2-1B-Instruct	1054	1047				8529	Meta	Llama 3.2
Step-1o-Vision-32k (highres)			1183				StepFun	Proprietary
Qwen2.5-VL-72B-Instruct			1166				Alibaba	Qwen
Pixtral-Large-2411			1153				Mistral	MRL
Qwen-VL-Max-1119			1127				Alibaba	Proprietary
Qwen2-VL-72b-Instruct			1111				Alibaba	Qwen
Step-1V-32K			1110				StepFun	Proprietary
Molmo-72B-0924			1075				AI2	Apache 2.0
Pixtral-12B-2409			1072				Mistral	Apache 2.0
Llama-3.2-90B-Vision-Instruct			1068				Meta	Llama 3.2
InternVL2-26B			1067				OpenGVLab	MIT
Qwen2-VL-7B-Instruct			1052				Aliaba	Apache 2.0
Yi-Vision			1045				01 AI	Proprietary
Llama-3.2-11B-Vision-Instruct			1031				Meta	Llama 3.2

If you want to see more models, please help us add them.

💻 Code: The Arena Elo ratings are computed by this notebook. The MT-bench scores (single-answer grading on a scale of 10) are computed by fastchat.llm_judge. The MMLU scores are computed by InstructEval. Higher values are better for all benchmarks. Empty cells mean not available. The latest and detailed leaderboard is here.

More Statistics for Chatbot Arena

🔗 Arena Statistics

Transition from online Elo rating system to Bradley-Terry model

We adopted the Elo rating system for ranking models since the launch of the Arena. It has been useful to transform pairwise human preference to Elo ratings that serve as a predictor of winrate between models. Specifically, if player A has a rating of R_A and player B a rating of R_B, the probability of player A winning is

ELO rating has been used to rank chess players by the international community for over 60 years. Standard Elo rating systems assume a player’s performance changes overtime. So an online algorithm is needed to capture such dynamics, meaning recent games should weigh more than older games. Specifically, after each game, a player’s rating is updated according to the difference between predicted outcome and actual outcome.

This algorithm has two distinct features:

1. It can be computed asynchronously by players around the world.
2. It allows for players performance to change dynamically – it does not assume a fixed unknown value for the players rating.

This ability to adapt is determined by the parameter K which controls the magnitude of rating changes that can affect the overall result. A larger K essentially put more weight on the recent games, which may make sense for new players whose performance improves quickly. However as players become more senior and their performance “converges” then a smaller value of K is more appropriate. As a result, USCF adopted K based on the number of games and tournaments completed by the player (reference). That is, the Elo rating of a senior player changes slower than a new player.

When we launched the Arena, we noticed considerable variability in the ratings using the classic online algorithm. We tried to tune the K to be sufficiently stable while also allowing new models to move up quickly in the leaderboard. We ultimately decided to adopt a bootstrap-like technique to shuffle the data and sample Elo scores from 1000 permutations of the online plays. You can find the details in this notebook. This provided consistent stable scores and allowed us to incorporate new models quickly. This is also observed in a recent work by Cohere. However, we used the same samples to estimate confidence intervals which were therefore too wide (effectively CI’s for the original online Elo estimates).

In the context of LLM ranking, there are two important differences from the classic Elo chess ranking system. First, we have access to the entire history of all games for all models and so we don’t need a decentralized algorithm. Second, most models are static (we have access to the weights) and so we don’t expect their performance to change. However, it is worth noting that the hosted proprietary models may not be static and their behavior can change without notice. We try our best to pin specific model API versions if possible.

To improve the quality of our rankings and their confidence estimates, we are adopting another widely used rating system called the Bradley–Terry (BT) model. This model actually is the maximum likelihood (MLE) estimate of the underlying Elo model assuming a fixed but unknown pairwise win-rate. Similar to Elo rating, BT model is also based on pairwise comparison to derive ratings of players to estimate win rate between each other. The core difference between BT model vs the online Elo system is the assumption that player’s performance does not change (i.e., game order does not matter) and the computation takes place in a centralized fashion.