Text Arena

View rankings across various LLMs on their versatility, linguistic precision, and cultural context across text

Last Updated

Jan 27, 2026

Total Votes

5,135,830

Total Models

299

/

Style Control

	Rank Spread
1	1◄─►1	gemini-3-pro	1488	±5	31,624	Google	Proprietary
2	2◄─►5	grok-4.1-thinking	1476	±5	31,545	xAI	Proprietary
3	2◄─►7	gemini-3-flash	1471	±5	18,772	Google	Proprietary
4	2◄─►7	claude-opus-4-5-20251101-thinking-32k	1468	±5	23,402	Anthropic	Proprietary
5	2◄─►8	claude-opus-4-5-20251101	1467	±5	28,091	Anthropic	Proprietary
6	3◄─►8	grok-4.1	1466	±5	35,971	xAI	Proprietary
7	3◄─►9	gemini-3-flash (thinking-minimal)	1464	±6	13,318	Google	Proprietary
8	5◄─►13	gpt-5.1-high	1458	±5	27,690	OpenAI	Proprietary
9	7◄─►20	ernie-5.0-0110	1454	±8	7,732	Baidu	Proprietary
10	8◄─►19	claude-sonnet-4-5-20250929-thinking-32k	1451	±4	41,942	Anthropic	Proprietary
11	9◄─►19	gemini-2.5-pro	1450	±3	91,130	Google	Proprietary
12	8◄─►20	claude-sonnet-4-5-20250929	1450	±4	39,442	Anthropic	Proprietary
13	8◄─►21	ernie-5.0-preview-1203	1449	±7	9,785	Baidu	Proprietary
14	9◄─►21	claude-opus-4-1-20250805-thinking-16k	1448	±4	49,980	Anthropic	Proprietary
15	8◄─►26	kimi-k2.5-thinking	1446	±11	2,551	Moonshot	Modified MIT
16	9◄─►21	claude-opus-4-1-20250805	1445	±3	71,058	Anthropic	Proprietary
17	9◄─►24	gpt-4.5-preview-2025-02-27	1444	±6	14,549	OpenAI	Proprietary
18	13◄─►22	chatgpt-4o-latest-20250326	1442	±3	78,363	OpenAI	Proprietary
19	9◄─►26	glm-4.7	1441	±6	12,024	Z.ai	MIT
20	9◄─►27	gpt-5.2-high	1441	±7	12,236	OpenAI	Proprietary
21	11◄─►27	gpt-5.2	1439	±7	8,751	OpenAI	Proprietary
22	16◄─►27	gpt-5.1	1436	±5	29,739	OpenAI	Proprietary
23	17◄─►29	gpt-5-high	1435	±5	32,655	OpenAI	Proprietary
24	17◄─►32	qwen3-max-preview	1434	±5	27,859	Alibaba	Proprietary
25	18◄─►32	o3-2025-04-16	1433	±4	61,366	OpenAI	Proprietary
26	18◄─►35	grok-4-1-fast-reasoning	1431	±5	24,679	xAI	Proprietary
27	20◄─►37	kimi-k2-thinking-turbo	1430	±4	29,405	Moonshot	Modified MIT
28	23◄─►45	gpt-5-chat	1426	±4	31,851	OpenAI	Proprietary
29	23◄─►47	qwen3-max-2025-09-23	1425	±6	9,222	Alibaba	Proprietary
30	26◄─►46	glm-4.6	1425	±4	35,379	Z.ai	MIT
31	26◄─►46	claude-opus-4-20250514-thinking-16k	1424	±4	37,990	Anthropic	Proprietary
32	24◄─►49	deepseek-v3.2-exp	1423	±6	11,776	DeepSeek	MIT
33	24◄─►49	deepseek-v3.2-exp-thinking	1423	±7	9,001	DeepSeek	MIT
34	24◄─►51	grok-4-fast-chat	1422	±8	6,995	xAI	Proprietary
35	28◄─►47	qwen3-235b-a22b-instruct-2507	1422	±3	65,460	Alibaba	Apache 2.0
36	28◄─►51	deepseek-v3.2-thinking	1420	±5	19,145	DeepSeek	MIT
37	28◄─►52	deepseek-v3.2	1419	±5	23,984	DeepSeek	MIT
38	28◄─►53	deepseek-r1-0528	1418	±6	19,290	DeepSeek	MIT
39	28◄─►54	kimi-k2-0905-preview	1418	±6	11,978	Moonshot	Modified MIT
40	26◄─►56	ernie-5.0-preview-1022	1417	±9	4,630	Baidu	Proprietary
41	28◄─►53	kimi-k2-0711-preview	1417	±5	28,672	Moonshot	Modified MIT
42	28◄─►54	deepseek-v3.1	1417	±6	15,305	DeepSeek	MIT
43	28◄─►54	deepseek-v3.1-thinking	1417	±7	11,988	DeepSeek	MIT
44	27◄─►58	deepseek-v3.1-terminus	1416	±10	3,763	DeepSeek	MIT
45	27◄─►59	deepseek-v3.1-terminus-thinking	1416	±10	3,547	DeepSeek	MIT
46	29◄─►56	qwen3-vl-235b-a22b-instruct	1415	±6	11,684	Alibaba	Apache 2.0
47	31◄─►54	mistral-large-3	1415	±5	20,189	Mistral	Apache 2.0
48	33◄─►54	claude-opus-4-20250514	1414	±4	45,575	Anthropic	Proprietary
49	33◄─►54	gpt-4.1-2025-04-14	1413	±4	52,224	OpenAI	Proprietary
50	35◄─►56	mistral-medium-2508	1412	±3	59,305	Mistral	Proprietary
51	35◄─►58	grok-3-preview-02-24	1411	±4	33,982	xAI	Proprietary
52	38◄─►59	grok-4-0709	1410	±4	42,171	xAI	Proprietary
53	37◄─►62	glm-4.5	1409	±5	24,806	Z.ai	MIT
54	40◄─►58	gemini-2.5-flash	1409	±3	90,416	Google	Proprietary
55	46◄─►67	gemini-2.5-flash-preview-09-2025	1405	±4	32,897	Google	Proprietary
56	46◄─►68	grok-4-fast-reasoning	1404	±5	18,654	xAI	Proprietary
57	49◄─►68	claude-haiku-4-5-20251001	1403	±4	40,529	Anthropic	Proprietary
58	49◄─►68	o1-2024-12-17	1402	±4	27,822	OpenAI	Proprietary
59	54◄─►69	qwen3-235b-a22b-no-thinking	1401	±4	39,556	Alibaba	Apache 2.0
60	52◄─►70	qwen3-next-80b-a3b-instruct	1401	±5	22,875	Alibaba	Apache 2.0
61	54◄─►70	claude-sonnet-4-20250514-thinking-32k	1400	±4	36,218	Anthropic	Proprietary
62	54◄─►76	longcat-flash-chat	1399	±6	11,557	Meituan	MIT
63	55◄─►76	qwen3-235b-a22b-thinking-2507	1398	±6	9,249	Alibaba	Apache 2.0
64	55◄─►76	deepseek-r1	1398	±5	18,537	DeepSeek	MIT
65	55◄─►83	amazon-nova-experimental-chat-12-10	1395	±10	3,718	Amazon	Proprietary
66	55◄─►82	qwen3-vl-235b-a22b-thinking	1395	±7	7,974	Alibaba	Apache 2.0
67	56◄─►80	mimo-v2-flash (non-thinking)	1395	±6	12,764	Xiaomi	MIT
68	59◄─►77	deepseek-v3-0324	1394	±4	46,701	DeepSeek	MIT
69	54◄─►84	hunyuan-vision-1.5-thinking	1393	±12	2,223	Tencent	Proprietary
70	62◄─►82	o4-mini-2025-04-16	1391	±4	46,685	OpenAI	Proprietary
71	60◄─►83	mai-1-preview	1391	±5	18,136	Microsoft AI	Proprietary
72	62◄─►83	gpt-5-mini-high	1391	±5	27,177	OpenAI	Proprietary
73	62◄─►83	claude-sonnet-4-20250514	1390	±4	41,655	Anthropic	Proprietary
74	62◄─►83	claude-3-7-sonnet-20250219-thinking-32k	1389	±4	39,889	Anthropic	Proprietary
75	62◄─►83	o1-preview	1388	±5	31,120	OpenAI	Proprietary
76	62◄─►86	hunyuan-t1-20250711	1387	±9	4,800	Tencent	Proprietary
77	65◄─►84	qwen3-coder-480b-a35b-instruct	1386	±5	26,650	Alibaba	Apache 2.0
78	66◄─►84	mistral-medium-2505	1384	±5	34,550	Mistral	Proprietary
79	67◄─►86	qwen3-30b-a3b-instruct-2507	1383	±5	24,095	Alibaba	Apache 2.0
80	66◄─►89	hunyuan-turbos-20250416	1382	±6	11,051	Tencent	Proprietary
81	67◄─►89	minimax-m2.1-preview	1382	±6	12,276	MiniMax	MIT
82	69◄─►87	gpt-4.1-mini-2025-04-14	1382	±4	40,535	OpenAI	Proprietary
83	75◄─►90	gemini-2.5-flash-lite-preview-09-2025-no-thinking	1379	±4	43,595	Google	Proprietary
84	66◄─►96	glm-4.6v	1378	±11	2,823	Z.ai	MIT
85	78◄─►93	gemini-2.5-flash-lite-preview-06-17-thinking	1375	±4	33,937	Google	Proprietary
86	78◄─►93	qwen3-235b-a22b	1374	±5	27,158	Alibaba	Apache 2.0
87	80◄─►93	qwen2.5-max	1374	±4	33,288	Alibaba	Proprietary
88	81◄─►93	claude-3-5-sonnet-20241022	1373	±3	89,456	Anthropic	Proprietary
89	81◄─►95	claude-3-7-sonnet-20250219	1372	±4	44,435	Anthropic	Proprietary
90	83◄─►96	glm-4.5-air	1371	±4	31,399	Z.ai	MIT
91	84◄─►99	qwen3-next-80b-a3b-thinking	1369	±6	13,870	Alibaba	Apache 2.0
92	84◄─►99	minimax-m1	1367	±4	36,845	MiniMax	Apache 2.0
93	84◄─►103	amazon-nova-experimental-chat-11-10	1365	±6	12,863	Amazon	Proprietary
94	88◄─►99	gemma-3-27b-it	1365	±4	48,690	Google	Gemma
95	88◄─►103	o3-mini-high	1364	±5	18,584	OpenAI	Proprietary
96	89◄─►107	grok-3-mini-high	1362	±5	17,525	xAI	Proprietary
97	91◄─►107	gemini-2.0-flash-001	1361	±4	44,815	Google	Proprietary
98	91◄─►113	deepseek-v3	1358	±5	21,788	DeepSeek	DeepSeek
99	94◄─►118	grok-3-mini-beta	1357	±5	23,765	xAI	Proprietary
100	94◄─►119	mistral-small-2506	1356	±5	18,344	Mistral	Apache 2.0
101	91◄─►121	intellect-3	1356	±8	5,326	Prime Intellect	MIT
102	96◄─►120	gpt-oss-120b	1354	±4	30,980	OpenAI	Apache 2.0
103	96◄─►121	gemini-2.0-flash-lite-preview-02-05	1353	±4	24,951	Google	Proprietary
104	94◄─►123	glm-4.5v	1353	±8	4,978	Z.ai	MIT
105	98◄─►121	command-a-03-2025	1353	±3	57,447	Cohere	CC-BY-NC-4.0
106	98◄─►121	gemini-1.5-pro-002	1352	±3	55,607	Google	Proprietary
107	94◄─►133	hunyuan-turbos-20250226	1349	±12	2,226	Tencent	Proprietary
108	99◄─►123	o3-mini	1348	±3	58,647	OpenAI	Proprietary
109	98◄─►125	amazon-nova-experimental-chat-10-20	1348	±6	11,436	Amazon	Proprietary
110	96◄─►134	llama-3.1-nemotron-ultra-253b-v1	1347	±12	2,546	Nvidia	Nvidia Open Model
111	96◄─►134	amazon-nova-experimental-chat-10-09	1347	±11	2,893	Amazon	Proprietary
112	98◄─►131	minimax-m2	1347	±8	6,749	MiniMax	Apache 2.0
113	98◄─►133	qwen3-32b	1347	±9	3,932	Alibaba	Apache 2.0
114	99◄─►130	ling-flash-2.0	1346	±7	7,045	Ant Group	MIT
115	99◄─►132	step-3	1346	±7	6,594	StepFun	Apache 2.0
116	98◄─►132	qwen-plus-0125	1346	±8	5,823	Alibaba	Proprietary
117	102◄─►123	gpt-4o-2024-05-13	1346	±3	112,863	OpenAI	Proprietary
118	99◄─►134	glm-4-plus-0111	1343	±8	5,760	Zhipu	Proprietary
119	106◄─►128	claude-3-5-sonnet-20240620	1343	±3	82,417	Anthropic	Proprietary
120	100◄─►138	gemma-3-12b-it	1342	±10	3,829	Google	Gemma
121	101◄─►139	nvidia-llama-3.3-nemotron-super-49b-v1.5	1341	±10	3,430	Nvidia	Nvidia Open
122	99◄─►140	hunyuan-turbo-0110	1340	±12	2,295	Tencent	Proprietary
123	106◄─►139	gpt-5-nano-high	1338	±7	8,404	OpenAI	Proprietary
124	110◄─►136	o1-mini	1337	±4	51,986	OpenAI	Proprietary
125	109◄─►139	nova-2-lite	1336	±6	12,275	Amazon	Proprietary
126	111◄─►137	llama-3.1-405b-instruct-bf16	1336	±4	41,392	Meta	Llama 3.1 Community
127	110◄─►139	qwq-32b	1336	±4	26,127	Alibaba	Apache 2.0
128	111◄─►139	gpt-4o-2024-08-06	1335	±4	45,498	OpenAI	Proprietary
129	110◄─►139	gemini-advanced-0514	1335	±5	50,142	Google	Proprietary
130	113◄─►139	grok-2-2024-08-13	1335	±4	63,495	xAI	Proprietary
131	115◄─►139	llama-3.1-405b-instruct-fp8	1334	±4	59,655	Meta	Llama 3.1 Community
132	109◄─►147	step-2-16k-exp-202412	1334	±9	4,829	StepFun	Proprietary
133	120◄─►152	yi-lightning	1329	±5	27,340	01 AI	Proprietary
134	122◄─►152	llama-4-maverick-17b-128e-instruct	1328	±4	41,140	Meta	Llama 4
135	121◄─►153	qwen3-30b-a3b	1328	±5	27,430	Alibaba	Apache 2.0
136	112◄─►161	llama-3.3-nemotron-49b-super-v1	1327	±12	2,230	Nvidia	Nvidia
137	117◄─►161	hunyuan-large-2025-02-10	1327	±10	3,738	Tencent	Proprietary
138	120◄─►158	olmo-3.1-32b-instruct	1326	±8	7,721	Allen AI	Apache 2.0
139	132◄─►157	gpt-4-turbo-2024-04-09	1325	±4	98,130	OpenAI	Proprietary
140	132◄─►157	claude-3-5-haiku-20241022	1324	±3	71,184	Anthropic	Proprietary
141	132◄─►157	gemini-1.5-pro-001	1323	±4	79,132	Google	Proprietary
142	123◄─►161	deepseek-v2.5-1210	1323	±8	6,793	DeepSeek	DeepSeek
143	132◄─►157	claude-3-opus-20240229	1323	±3	194,904	Anthropic	Proprietary
144	132◄─►159	llama-4-scout-17b-16e-instruct	1323	±5	31,233	Meta	Llama
145	131◄─►162	gpt-4.1-nano-2025-04-14	1322	±8	6,107	OpenAI	Proprietary
146	132◄─►161	step-1o-turbo-202506	1321	±7	9,688	StepFun	Proprietary
147	132◄─►163	ring-flash-2.0	1320	±7	7,198	Ant Group	MIT
148	135◄─►161	llama-3.3-70b-instruct	1320	±3	55,563	Meta	Llama-3.3
149	133◄─►161	glm-4-plus	1319	±5	26,134	Zhipu AI	Proprietary
150	133◄─►162	gemma-3n-e4b-it	1319	±5	23,347	Google	Gemma
151	133◄─►163	qwen-max-0919	1318	±6	16,479	Alibaba	Qwen
152	136◄─►162	gpt-4o-mini-2024-07-18	1318	±3	68,801	OpenAI	Proprietary
153	133◄─►168	gpt-oss-20b	1318	±6	10,807	OpenAI	Apache 2.0
154	136◄─►168	nvidia-nemotron-3-nano-30b-a3b-bf16	1317	±6	15,395	Nvidia	NVIDIA Open Model
155	136◄─►170	qwen2.5-plus-1127	1315	±6	10,179	Alibaba	Proprietary
156	140◄─►168	athene-v2-chat	1314	±4	24,746	NexusFlow	NexusFlow
157	141◄─►168	mistral-large-2407	1314	±4	45,460	Mistral	Mistral Research
158	142◄─►168	gpt-4-1106-preview	1314	±4	100,107	OpenAI	Proprietary
159	142◄─►168	gpt-4-0125-preview	1314	±4	93,439	OpenAI	Proprietary
160	136◄─►173	hunyuan-standard-2025-02-10	1312	±10	3,905	Tencent	Proprietary
161	133◄─►177	mercury	1310	±14	1,921	Inception AI	Proprietary
162	148◄─►172	gemini-1.5-flash-002	1310	±4	34,909	Google	Proprietary
163	153◄─►173	grok-2-mini-2024-08-13	1308	±4	52,574	xAI	Proprietary
164	153◄─►173	deepseek-v2.5	1307	±5	24,574	DeepSeek	DeepSeek
165	153◄─►173	athene-70b-0725	1306	±6	19,622	NexusFlow	CC-BY-NC-4.0
166	153◄─►173	magistral-medium-2506	1305	±6	12,062	Mistral	Proprietary
167	159◄─►173	mistral-large-2411	1305	±4	28,081	Mistral	MRL
168	151◄─►177	olmo-3-32b-think	1305	±8	5,895	Allen AI	Apache 2.0
169	159◄─►173	mistral-small-3.1-24b-instruct-2503	1305	±4	34,135	Mistral	Apache 2.0
170	153◄─►180	gemma-3-4b-it	1303	±9	4,177	Google	Gemma
171	160◄─►173	qwen2.5-72b-instruct	1303	±4	39,409	Alibaba	Qwen
172	160◄─►183	llama-3.1-nemotron-70b-instruct	1299	±8	7,136	Nvidia	Llama 3.1
173	161◄─►184	hunyuan-large-vision	1296	±9	5,602	Tencent	Proprietary
174	169◄─►184	llama-3.1-70b-instruct	1294	±4	55,234	Meta	Llama 3.1 Community
175	171◄─►186	amazon-nova-pro-v1.0	1290	±5	24,753	Amazon	Proprietary
176	171◄─►188	jamba-1.5-large	1289	±7	8,659	AI21 Labs	Jamba Open
177	169◄─►190	ibm-granite-h-small	1289	±8	5,677	IBM	Apache 2.0
178	172◄─►186	gemma-2-27b-it	1288	±3	75,764	Google	Gemma license
179	171◄─►188	reka-core-20240904	1288	±7	7,309	Reka AI	Proprietary
180	172◄─►188	gpt-4-0314	1288	±5	54,167	OpenAI	Proprietary
181	169◄─►194	llama-3.1-nemotron-51b-instruct	1287	±10	3,749	Nvidia	Llama 3.1
182	169◄─►194	llama-3.1-tulu-3-70b	1287	±10	2,846	Ai2	Llama 3.1
183	173◄─►188	gemini-1.5-flash-001	1286	±4	62,823	Google	Proprietary
184	172◄─►194	olmo-3.1-32b-think	1285	±7	8,497	Allen AI	Apache 2.0
185	175◄─►194	claude-3-sonnet-20240229	1282	±4	109,289	Anthropic	Proprietary
186	175◄─►194	gemma-2-9b-it-simpo	1280	±7	10,069	Princeton	MIT
187	177◄─►194	nemotron-4-340b-instruct	1278	±5	19,661	Nvidia	NVIDIA Open Model
188	177◄─►196	command-r-plus-08-2024	1277	±7	9,869	Cohere	CC-BY-NC-4.0
189	182◄─►194	llama-3-70b-instruct	1277	±4	156,880	Meta	Llama 3 Community
190	181◄─►195	gpt-4-0613	1276	±4	88,721	OpenAI	Proprietary
191	182◄─►197	mistral-small-24b-instruct-2501	1274	±6	14,677	Mistral	Apache 2.0
192	181◄─►199	glm-4-0520	1274	±7	9,788	Zhipu AI	Proprietary
193	182◄─►200	reka-flash-20240904	1273	±7	7,537	Reka AI	Proprietary
194	182◄─►203	qwen2.5-coder-32b-instruct	1271	±8	5,430	Alibaba	Apache 2.0
195	189◄─►203	c4ai-aya-expanse-32b	1267	±5	27,123	Cohere	CC-BY-NC-4.0
196	191◄─►203	gemma-2-9b-it	1266	±4	54,615	Google	Gemma license
197	190◄─►204	deepseek-coder-v2	1265	±6	15,147	DeepSeek	DeepSeek License
198	192◄─►204	command-r-plus	1263	±4	77,556	Cohere	CC-BY-NC-4.0
199	192◄─►204	qwen2-72b-instruct	1263	±5	37,325	Alibaba	Qianwen LICENSE
200	194◄─►204	claude-3-haiku-20240307	1262	±4	117,705	Anthropic	Proprietary
201	193◄─►205	amazon-nova-lite-v1.0	1261	±5	19,376	Amazon	Proprietary
202	194◄─►205	gemini-1.5-flash-8b-001	1259	±4	35,556	Google	Proprietary
203	197◄─►205	phi-4	1256	±5	24,126	Microsoft	MIT
204	194◄─►211	olmo-2-0325-32b-instruct	1253	±11	3,335	Allen AI	Apache-2.0
205	201◄─►210	command-r-08-2024	1251	±7	10,141	Cohere	CC-BY-NC-4.0
206	204◄─►214	mistral-large-2402	1243	±5	62,437	Mistral	Proprietary
207	204◄─►214	amazon-nova-micro-v1.0	1241	±5	19,355	Amazon	Proprietary
208	204◄─►218	jamba-1.5-mini	1239	±7	8,854	AI21 Labs	Jamba Open
209	204◄─►222	ministral-8b-2410	1237	±9	4,780	Mistral	MRL
210	205◄─►222	gemini-pro-dev-api	1236	±7	18,352	Google	Proprietary
211	206◄─►222	qwen1.5-110b-chat	1235	±6	26,191	Alibaba	Qianwen LICENSE
212	206◄─►223	reka-flash-21b-20240226-online	1234	±7	15,451	Reka AI	Proprietary
213	206◄─►222	qwen1.5-72b-chat	1234	±5	39,296	Alibaba	Qianwen LICENSE
214	204◄─►224	hunyuan-standard-256k	1233	±12	2,729	Tencent	Proprietary
215	208◄─►223	mixtral-8x22b-instruct-v0.1	1231	±5	51,417	Mistral	Apache 2.0
216	208◄─►224	command-r	1228	±5	54,038	Cohere	CC-BY-NC-4.0
217	208◄─►224	reka-flash-21b-20240226	1227	±6	24,806	Reka AI	Proprietary
218	209◄─►225	gpt-3.5-turbo-0125	1225	±5	66,191	OpenAI	Proprietary
219	209◄─►226	mistral-medium	1224	±5	34,552	Mistral	Proprietary
220	213◄─►225	llama-3-8b-instruct	1224	±4	104,636	Meta	Llama 3 Community
221	209◄─►226	c4ai-aya-expanse-8b	1223	±7	9,827	Cohere	CC-BY-NC-4.0
222	208◄─►229	gemini-pro	1223	±12	6,390	Google	Proprietary
223	209◄─►229	llama-3.1-tulu-3-8b	1221	±11	2,895	Ai2	Llama 3.1
224	220◄─►229	yi-1.5-34b-chat	1214	±5	24,142	01 AI	Apache-2.0
225	215◄─►230	zephyr-orpo-141b-A35b-v0.1	1214	±11	4,653	HuggingFace	Apache 2.0
226	222◄─►229	llama-3.1-8b-instruct	1212	±4	49,605	Meta	Llama 3.1 Community
227	218◄─►235	granite-3.1-8b-instruct	1209	±11	3,092	IBM	Apache 2.0
228	222◄─►235	qwen1.5-32b-chat	1205	±6	21,744	Alibaba	Qianwen LICENSE
229	222◄─►237	gpt-3.5-turbo-1106	1204	±9	16,616	OpenAI	Proprietary
230	226◄─►237	phi-3-medium-4k-instruct	1199	±5	25,055	Microsoft	MIT
231	227◄─►237	gemma-2-2b-it	1199	±4	46,618	Google	Gemma license
232	227◄─►237	mixtral-8x7b-instruct-v0.1	1198	±4	73,505	Mistral	Apache 2.0
233	227◄─►242	dbrx-instruct-preview	1196	±6	32,196	Databricks	DBRX LICENSE
234	227◄─►246	internlm2_5-20b-chat	1192	±7	9,902	InternLM	Other
235	227◄─►246	qwen1.5-14b-chat	1192	±7	17,841	Alibaba	Qianwen LICENSE
236	229◄─►252	wizardlm-70b	1186	±9	8,214	Microsoft	Llama 2 Community
237	229◄─►253	deepseek-llm-67b-chat	1185	±12	4,933	DeepSeek	DeepSeek License
238	233◄─►249	yi-34b-chat	1185	±7	15,483	01 AI	Yi License
239	233◄─►252	openchat-3.5-0106	1183	±8	12,636	OpenChat	Apache-2.0
240	233◄─►253	openchat-3.5	1183	±10	7,967	OpenChat	Apache-2.0
241	233◄─►253	granite-3.0-8b-instruct	1183	±9	6,643	IBM	Apache 2.0
242	234◄─►253	gemma-1.1-7b-it	1181	±6	23,893	Google	Gemma license
243	234◄─►253	snowflake-arctic-instruct	1181	±6	32,836	Snowflake	Apache 2.0
244	233◄─►255	granite-3.1-2b-instruct	1180	±11	3,191	IBM	Apache 2.0
245	234◄─►253	tulu-2-dpo-70b	1179	±10	6,534	AllenAI/UW	AI2 ImpACT Low-risk
246	234◄─►257	openhermes-2.5-mistral-7b	1176	±10	5,006	NousResearch	Apache-2.0
247	236◄─►256	vicuna-33b	1174	±6	22,479	LMSYS	Non-commercial
248	236◄─►259	starling-lm-7b-beta	1172	±7	16,057	Nexusflow	Apache-2.0
249	236◄─►257	phi-3-small-8k-instruct	1172	±6	17,763	Microsoft	MIT
250	237◄─►257	llama-2-70b-chat	1172	±5	38,491	Meta	Llama 2 Community
251	237◄─►260	starling-lm-7b-alpha	1168	±8	10,224	UC Berkeley	CC-BY-NC-4.0
252	239◄─►260	llama-3.2-3b-instruct	1167	±8	7,936	Meta	Llama 3.2
253	237◄─►263	nous-hermes-2-mixtral-8x7b-dpo	1166	±12	3,776	NousResearch	Apache-2.0
254	245◄─►269	qwq-32b-preview	1158	±12	3,233	Alibaba	Apache 2.0
255	250◄─►267	granite-3.0-2b-instruct	1157	±8	6,837	IBM	Apache 2.0
256	245◄─►270	llama2-70b-steerlm-chat	1156	±13	3,584	Nvidia	Llama 2 Community
257	247◄─►273	solar-10.7b-instruct-v1.0	1153	±13	4,155	Upstage AI	CC-BY-NC-4.0
258	246◄─►275	dolphin-2.2.1-mistral-7b	1153	±15	1,679	Cognitive Computations	Apache-2.0
259	251◄─►273	mpt-30b-chat	1151	±12	2,571	MosaicML	CC-BY-NC-SA-4.0
260	253◄─►270	mistral-7b-instruct-v0.2	1151	±7	19,402	Mistral	Apache-2.0
261	253◄─►272	wizardlm-13b	1150	±9	7,046	Microsoft	Llama 2 Community
262	250◄─►277	falcon-180b-chat	1148	±17	1,295	TII	Falcon-180B TII License
263	253◄─►276	qwen1.5-7b-chat	1145	±10	4,735	Alibaba	Qianwen LICENSE
264	254◄─►275	phi-3-mini-4k-instruct-june-2024	1144	±6	12,296	Microsoft	MIT
265	254◄─►276	llama-2-13b-chat	1142	±7	19,171	Meta	Llama 2 Community
266	254◄─►276	vicuna-13b	1142	±7	19,366	LMSYS	Llama 2 Community
267	254◄─►278	qwen-14b-chat	1140	±11	4,964	Alibaba	Qianwen LICENSE
268	255◄─►278	palm-2	1138	±9	8,554	Google	Proprietary
269	255◄─►278	codellama-34b-instruct	1138	±9	7,363	Meta	Llama 2 Community
270	256◄─►278	gemma-7b-it	1137	±10	8,925	Google	Gemma license
271	258◄─►279	zephyr-7b-beta	1132	±9	11,116	HuggingFace	MIT
272	261◄─►279	phi-3-mini-128k-instruct	1130	±7	20,691	Microsoft	MIT
273	263◄─►279	phi-3-mini-4k-instruct	1129	±6	20,115	Microsoft	MIT
274	259◄─►282	guanaco-33b	1128	±12	2,921	UW	Non-commercial
275	258◄─►283	zephyr-7b-alpha	1128	±16	1,785	HuggingFace	MIT
276	266◄─►283	stripedhyena-nous-7b	1122	±11	5,184	Together AI	Apache 2.0
277	261◄─►284	codellama-70b-instruct	1120	±18	1,143	Meta	Llama 2 Community
278	271◄─►283	vicuna-7b	1116	±9	6,923	LMSYS	Llama 2 Community
279	267◄─►284	smollm2-1.7b-instruct	1115	±14	2,201	HuggingFace	Apache 2.0
280	274◄─►283	gemma-1.1-2b-it	1115	±8	10,853	Google	Gemma license
281	274◄─►283	llama-3.2-1b-instruct	1112	±8	8,045	Meta	Llama 3.2
282	274◄─►284	mistral-7b-instruct	1111	±9	8,977	Mistral	Apache 2.0
283	275◄─►284	llama-2-7b-chat	1109	±7	14,148	Meta	Llama 2 Community
284	280◄─►288	gemma-2b-it	1092	±12	4,779	Google	Gemma license
285	284◄─►287	qwen1.5-4b-chat	1091	±9	7,598	Alibaba	Qianwen LICENSE
286	284◄─►291	olmo-7b-instruct	1075	±11	6,329	Allen AI	Apache-2.0
287	285◄─►291	koala-13b	1071	±10	6,964	UC Berkeley	Non-commercial
288	286◄─►291	alpaca-13b	1069	±12	5,745	Stanford	Non-commercial
289	284◄─►292	gpt4all-13b-snoozy	1067	±15	1,743	Nomic AI	Non-commercial
290	286◄─►292	mpt-7b-chat	1063	±12	3,925	MosaicML	CC-BY-NC-SA-4.0
291	286◄─►292	chatglm3-6b	1057	±12	4,658	Tsinghua	Apache-2.0
292	289◄─►294	RWKV-4-Raven-14B	1042	±11	4,845	RWKV	Apache 2.0
293	292◄─►294	chatglm2-6b	1025	±14	2,657	Tsinghua	Apache-2.0
294	292◄─►294	oasst-pythia-12b	1023	±11	6,311	OpenAssistant	Apache 2.0
295	295◄─►298	chatglm-6b	997	±13	4,914	Tsinghua	Non-commercial
296	295◄─►298	fastchat-t5-3b	992	±12	4,203	LMSYS	Apache 2.0
297	295◄─►298	dolly-v2-12b	981	±14	3,412	Databricks	MIT
298	295◄─►299	llama-13b	973	±16	2,391	Meta	Non-commercial
299	298◄─►299	stablelm-tuned-alpha-7b	954	±13	3,287	Stability AI	CC-BY-NC-SA-4.0

Text Arena

Default Leaderboard Plots

Fraction of Model A Wins for All Non-tied A vs. B Battles

Confidence Intervals on Model Strength (via Bootstrapping)

Battle Count for Each Combination of Models (without Ties)

Average Win Rate Against All Other Models (Uniform Sampling and No Ties)