Text Arena

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

Last Updated

Dec 23, 2025

Total Votes

241,145

Total Models

261

/

Style Control

	Rank Spread
1	1◄─►9	gemini-3-pro	1529	±20	1,030	Google	Proprietary
2	1◄─►10	grok-4.1-thinking	1521	±20	1,048	xAI	Proprietary
3	1◄─►22	gemini-3-flash	1520	±36	290	Google	Proprietary
4	1◄─►23	grok-4.1	1501	±19	1,120	xAI	Proprietary
5	1◄─►33	claude-opus-4-5-20251101-thinking-32k	1497	±25	655	Anthropic	Proprietary
6	1◄─►55	ernie-5.0-preview-1203	1495	±44	204	Baidu	Proprietary
7	1◄─►38	gpt-5.1-high	1491	±22	882	OpenAI	Proprietary
8	1◄─►43	claude-opus-4-5-20251101	1487	±24	721	Anthropic	Proprietary
9	3◄─►29	o3-2025-04-16	1485	±12	3,395	OpenAI	Proprietary
10	3◄─►33	chatgpt-4o-latest-20250326	1482	±11	3,546	OpenAI	Proprietary
11	3◄─►38	claude-opus-4-1-20250805-thinking-16k	1482	±13	2,355	Anthropic	Proprietary
12	3◄─►46	claude-sonnet-4-5-20250929	1478	±17	1,319	Anthropic	Proprietary
13	1◄─►60	gemini-3-flash (thinking-minimal)	1478	±34	329	Google	Proprietary
14	2◄─►60	deepseek-v3.2-exp-thinking	1476	±29	412	DeepSeek AI	MIT
15	3◄─►50	qwen3-max-preview	1475	±17	1,432	Alibaba	Proprietary
16	3◄─►45	claude-opus-4-1-20250805	1474	±12	3,119	Anthropic	Proprietary
17	3◄─►58	gpt-5.1	1470	±21	913	OpenAI	Proprietary
18	5◄─►48	gemini-2.5-pro	1469	±10	4,714	Google	Proprietary
19	3◄─►54	claude-sonnet-4-5-20250929-thinking-32k	1469	±17	1,416	Anthropic	Proprietary
20	4◄─►55	gpt-5-high	1467	±16	1,761	OpenAI	Proprietary
21	5◄─►56	gpt-5-chat	1465	±15	1,658	OpenAI	Proprietary
22	3◄─►76	deepseek-v3.1-terminus-thinking	1462	±38	223	DeepSeek AI	MIT
23	3◄─►72	gpt-5.2-high	1462	±31	410	OpenAI	Proprietary
24	6◄─►58	qwen3-235b-a22b-instruct-2507	1461	±12	2,735	Alibaba	Apache 2.0
25	3◄─►75	grok-4-fast-chat	1456	±30	391	xAI	Proprietary
26	8◄─►62	kimi-k2-0711-preview	1456	±15	1,754	Moonshot	Modified MIT
27	5◄─►72	kimi-k2-0905-preview	1456	±23	734	Moonshot	Modified MIT
28	5◄─►73	gpt-4.5-preview-2025-02-27	1455	±24	590	OpenAI	Proprietary
29	10◄─►61	claude-opus-4-20250514	1454	±12	2,709	Anthropic	Proprietary
30	10◄─►64	claude-opus-4-20250514-thinking-16k	1453	±13	2,286	Anthropic	Proprietary
31	3◄─►87	deepseek-v3.1-terminus	1452	±39	255	DeepSeek AI	MIT
32	8◄─►70	glm-4.6	1452	±18	1,275	Z.ai	MIT
33	6◄─►74	deepseek-v3.1-thinking	1451	±23	710	DeepSeek	MIT
34	5◄─►79	deepseek-v3.2-thinking	1450	±27	502	DeepSeek AI	MIT
35	6◄─►79	deepseek-v3.2	1448	±26	575	DeepSeek AI	MIT
36	6◄─►82	mistral-large-3	1446	±28	507	Mistral	Apache 2.0
37	10◄─►76	kimi-k2-thinking-turbo	1446	±21	939	Moonshot	Modified MIT
38	14◄─►72	grok-4-0709	1445	±13	2,303	xAI	Proprietary
39	10◄─►81	grok-4-1-fast-reasoning	1443	±24	637	xAI	Proprietary
40	15◄─►74	grok-3-preview-02-24	1443	±15	1,699	xAI	Proprietary
41	5◄─►94	hunyuan-t1-20250711	1443	±39	259	Tencent	Proprietary
42	8◄─►87	qwen3-max-2025-09-23	1442	±28	428	Alibaba	Proprietary
43	8◄─►89	ernie-5.0-preview-1103	1441	±30	417	Baidu	Proprietary
44	13◄─►79	qwen3-next-80b-a3b-instruct	1441	±18	1,158	Alibaba	Apache 2.0
45	13◄─►80	deepseek-v3.1	1441	±20	942	DeepSeek	MIT
46	10◄─►87	longcat-flash-chat	1441	±24	648	Meituan	MIT
47	16◄─►74	gpt-4.1-2025-04-14	1440	±12	2,821	OpenAI	Proprietary
48	15◄─►74	mistral-medium-2508	1440	±13	2,524	Mistral	Proprietary
49	15◄─►79	deepseek-r1-0528	1440	±18	1,303	DeepSeek	MIT
50	14◄─►87	grok-4-fast-reasoning	1437	±21	807	xAI	Proprietary
51	11◄─►90	deepseek-v3.2-exp	1436	±26	548	DeepSeek AI	MIT
52	12◄─►90	qwen3-235b-a22b-thinking-2507	1436	±26	533	Alibaba	Apache 2.0
53	11◄─►90	qwen3-vl-235b-a22b-instruct	1435	±27	468	Alibaba	Apache 2.0
54	21◄─►80	claude-sonnet-4-20250514	1434	±13	2,445	Anthropic	Proprietary
55	19◄─►84	gemini-2.5-flash-preview-09-2025	1433	±16	1,533	Google	Proprietary
56	19◄─►87	glm-4.5	1432	±17	1,367	Z.ai	MIT
57	15◄─►91	hunyuan-turbos-20250416	1432	±25	617	Tencent	Proprietary
58	23◄─►81	deepseek-v3-0324	1431	±12	2,721	DeepSeek	MIT
59	8◄─►109	amazon-nova-experimental-chat-11-10	1429	±41	218	Amazon	Proprietary
60	21◄─►90	mai-1-preview	1429	±18	1,079	Microsoft AI	Proprietary
61	25◄─►87	claude-3-7-sonnet-20250219-thinking-32k	1428	±13	2,329	Anthropic	Proprietary
62	25◄─►87	qwen3-235b-a22b-no-thinking	1427	±13	2,410	Alibaba	Apache 2.0
63	26◄─►87	o4-mini-2025-04-16	1427	±13	2,563	OpenAI	Proprietary
64	26◄─►87	claude-sonnet-4-20250514-thinking-32k	1426	±13	2,215	Anthropic	Proprietary
65	27◄─►87	gemini-2.5-flash	1423	±10	4,478	Google	Proprietary
66	26◄─►94	claude-haiku-4-5-20251001	1420	±17	1,376	Anthropic	Proprietary
67	27◄─►92	mistral-medium-2505	1420	±14	2,044	Mistral	Proprietary
68	18◄─►110	glm-4-plus-0111	1418	±34	295	Zhipu	Proprietary
69	26◄─►100	deepseek-r1	1416	±20	830	DeepSeek	MIT
70	30◄─►98	qwen3-coder-480b-a35b-instruct	1414	±17	1,378	Alibaba	Apache 2.0
71	35◄─►95	minimax-m1	1413	±13	2,249	MiniMax	Apache 2.0
72	24◄─►110	qwen3-vl-235b-a22b-thinking	1413	±29	419	Alibaba	Apache 2.0
73	36◄─►103	gpt-5-mini-high	1409	±17	1,393	OpenAI	Proprietary
74	42◄─►101	gemini-2.5-flash-lite-preview-06-17-thinking	1408	±14	1,995	Google	Proprietary
75	38◄─►104	gemini-2.5-flash-lite-preview-09-2025-no-thinking	1407	±16	1,469	Google	Proprietary
76	46◄─►101	gemma-3-27b-it	1406	±13	2,413	Google	Gemma
77	38◄─►108	o1-2024-12-17	1406	±17	1,277	OpenAI	Proprietary
78	47◄─►103	gpt-4.1-mini-2025-04-14	1404	±13	2,378	OpenAI	Proprietary
79	48◄─►111	qwen3-30b-a3b-instruct-2507	1400	±17	1,329	Alibaba	Apache 2.0
80	48◄─►112	deepseek-v3	1399	±18	1,059	DeepSeek	DeepSeek
81	58◄─►112	glm-4.5-air	1397	±15	1,611	Z.ai	MIT
82	59◄─►112	qwen2.5-max	1396	±15	1,647	Alibaba	Proprietary
83	26◄─►139	hunyuan-large-2025-02-10	1396	±39	212	Tencent	Proprietary
84	65◄─►109	claude-3-5-sonnet-20241022	1396	±10	4,617	Anthropic	Proprietary
85	47◄─►120	qwen3-next-80b-a3b-thinking	1395	±23	709	Alibaba	Apache 2.0
86	26◄─►141	glm-4.5v	1395	±40	225	Z.ai	MIT
87	58◄─►115	mistral-small-2506	1394	±18	1,140	Mistral	Apache 2.0
88	64◄─►112	claude-3-7-sonnet-20250219	1394	±13	2,324	Anthropic	Proprietary
89	38◄─►136	qwen-plus-0125	1392	±32	315	Alibaba	Proprietary
90	63◄─►125	grok-3-mini-high	1389	±19	1,007	xAI	Proprietary
91	31◄─►149	nvidia-llama-3.3-nemotron-super-49b-v1.5	1389	±40	202	Nvidia	Nvidia Open
92	67◄─►125	qwen3-235b-a22b	1385	±15	1,590	Alibaba	Apache 2.0
93	69◄─►124	gemini-2.0-flash-001	1383	±13	2,370	Google	Proprietary
94	72◄─►124	command-a-03-2025	1381	±11	3,094	Cohere	CC-BY-NC-4.0
95	48◄─►151	step-2-16k-exp-202412	1380	±37	265	StepFun	Proprietary
96	44◄─►153	qwen3-32b	1380	±39	212	Alibaba	Apache 2.0
97	69◄─►138	gemini-2.0-flash-lite-preview-02-05	1377	±19	1,059	Google	Proprietary
98	75◄─►133	llama-3.1-405b-instruct-bf16	1376	±14	1,907	Meta	Llama 3.1 Community
99	75◄─►135	o1-preview	1375	±15	1,730	OpenAI	Proprietary
100	70◄─►144	o3-mini-high	1374	±21	825	OpenAI	Proprietary
101	75◄─►139	gpt-oss-120b	1373	±16	1,611	OpenAI	Apache 2.0
102	68◄─►153	gpt-5-nano-high	1370	±28	462	OpenAI	Proprietary
103	65◄─►160	hunyuan-standard-2025-02-10	1369	±38	247	Tencent	Proprietary
104	68◄─►156	step-3	1367	±32	341	StepFun	Apache 2.0
105	86◄─►145	gemini-1.5-pro-002	1364	±12	2,690	Google	Proprietary
106	85◄─►149	grok-3-mini-beta	1363	±16	1,484	xAI	Proprietary
107	86◄─►145	gpt-4o-2024-05-13	1363	±10	6,004	OpenAI	Proprietary
108	75◄─►156	gpt-oss-20b	1363	±27	558	OpenAI	Apache 2.0
109	85◄─►149	yi-lightning	1362	±16	1,454	01 AI	Proprietary
110	68◄─►161	minimax-m2	1362	±36	300	MiniMax	Apache 2.0
111	86◄─►146	grok-2-2024-08-13	1362	±11	3,231	xAI	Proprietary
112	87◄─►146	claude-3-5-sonnet-20240620	1361	±11	4,494	Anthropic	Proprietary
113	87◄─►151	llama-4-maverick-17b-128e-instruct	1358	±13	2,369	Meta	Llama 4
114	86◄─►152	qwq-32b	1358	±16	1,401	Alibaba	Apache 2.0
115	80◄─►160	amazon-nova-experimental-chat-10-20	1358	±26	585	Amazon	Proprietary
116	76◄─►161	ling-flash-2.0	1357	±31	363	Ant Group	MIT
117	72◄─►166	intellect-3	1357	±36	286	Prime Intellect	MIT
118	91◄─►150	llama-3.1-405b-instruct-fp8	1357	±12	3,084	Meta	Llama 3.1 Community
119	78◄─►161	nova-2-lite	1357	±28	477	Amazon	Proprietary
120	91◄─►151	claude-3-5-haiku-20241022	1354	±11	3,710	Anthropic	Proprietary
121	91◄─►153	gpt-4o-2024-08-06	1354	±13	2,194	OpenAI	Proprietary
122	91◄─►153	gemini-advanced-0514	1353	±14	2,703	Google	Proprietary
123	74◄─►173	gemma-3-4b-it	1352	±39	195	Google	Gemma
124	91◄─►153	llama-3.3-70b-instruct	1352	±12	2,790	Meta	Llama-3.3
125	92◄─►153	o3-mini	1351	±11	3,114	OpenAI	Proprietary
126	81◄─►170	gpt-4.1-nano-2025-04-14	1350	±32	302	OpenAI	Proprietary
127	85◄─►166	reka-core-20240904	1350	±28	394	Reka AI	Proprietary
128	92◄─►154	grok-2-mini-2024-08-13	1349	±12	2,638	xAI	Proprietary
129	91◄─►157	llama-4-scout-17b-16e-instruct	1349	±15	1,840	Meta	Llama
130	87◄─►162	magistral-medium-2506	1348	±24	787	Mistral	Proprietary
131	97◄─►154	claude-3-opus-20240229	1347	±9	10,362	Anthropic	Proprietary
132	91◄─►160	gemma-3n-e4b-it	1347	±17	1,389	Google	Gemma
133	94◄─►160	glm-4-plus	1345	±15	1,521	Zhipu AI	Proprietary
134	98◄─►160	o1-mini	1342	±12	2,658	OpenAI	Proprietary
135	86◄─►175	deepseek-v2.5-1210	1342	±32	291	DeepSeek	DeepSeek
136	101◄─►160	gpt-4o-mini-2024-07-18	1340	±11	3,472	OpenAI	Proprietary
137	89◄─►175	llama-3.1-nemotron-70b-instruct	1339	±31	393	Nvidia	Llama 3.1
138	98◄─►163	qwen3-30b-a3b	1339	±15	1,678	Alibaba	Apache 2.0
139	103◄─►161	gpt-4-turbo-2024-04-09	1339	±11	5,060	OpenAI	Proprietary
140	98◄─►166	athene-v2-chat	1338	±17	1,198	NexusFlow	NexusFlow
141	95◄─►173	step-1o-turbo-202506	1335	±23	659	StepFun	Proprietary
142	99◄─►171	athene-70b-0725	1335	±18	1,051	NexusFlow	CC-BY-NC-4.0
143	103◄─►167	gemini-1.5-flash-002	1335	±15	1,701	Google	Proprietary
144	93◄─►175	qwen2.5-plus-1127	1334	±26	474	Alibaba	Proprietary
145	107◄─►167	mistral-large-2407	1333	±13	2,451	Mistral	Mistral Research
146	91◄─►178	ring-flash-2.0	1332	±31	378	Ant Group	MIT
147	106◄─►173	amazon-nova-pro-v1.0	1329	±17	1,096	Amazon	Proprietary
148	119◄─►172	gpt-4-0125-preview	1326	±11	5,170	OpenAI	Proprietary
149	111◄─►175	mistral-large-2411	1325	±16	1,230	Mistral	MRL
150	117◄─►175	mistral-small-3.1-24b-instruct-2503	1325	±14	1,991	Mistral	Apache 2.0
151	110◄─►175	qwen-max-0919	1324	±18	940	Alibaba	Qwen
152	87◄─►196	gemma-3-12b-it	1324	±47	155	Google	Gemma
153	94◄─►193	olmo-3-32b-think	1321	±38	289	Allen AI	Apache 2.0
154	119◄─►176	deepseek-v2.5	1320	±17	1,244	DeepSeek	DeepSeek
155	122◄─►175	qwen2.5-72b-instruct	1320	±13	1,963	Alibaba	Qwen
156	122◄─►175	llama-3.1-70b-instruct	1320	±12	2,838	Meta	Llama 3.1 Community
157	97◄─►198	nvidia-nemotron-3-nano-30b-a3b-bf16	1315	±40	265	Nvidia	NVIDIA Open Model
158	132◄─►176	gpt-4-1106-preview	1314	±11	5,544	OpenAI	Proprietary
159	133◄─►179	claude-3-sonnet-20240229	1313	±11	5,957	Anthropic	Proprietary
160	122◄─►192	command-r-plus-08-2024	1309	±24	563	Cohere	CC-BY-NC-4.0
161	139◄─►183	llama-3-70b-instruct	1309	±10	8,283	Meta	Llama 3 Community
162	137◄─►184	gemini-1.5-pro-001	1309	±11	4,440	Google	Proprietary
163	121◄─►193	gemma-2-9b-it-simpo	1309	±26	516	Princeton	MIT
164	103◄─►202	ibm-granite-h-small	1307	±43	241	IBM	Apache 2.0
165	145◄─►189	gemma-2-27b-it	1301	±10	3,882	Google	Gemma license
166	141◄─►193	c4ai-aya-expanse-32b	1299	±16	1,359	Cohere	CC-BY-NC-4.0
167	140◄─►194	nemotron-4-340b-instruct	1299	±18	1,070	Nvidia	NVIDIA Open Model
168	145◄─►192	command-r-plus	1298	±12	4,205	Cohere	CC-BY-NC-4.0
169	134◄─►199	command-r-08-2024	1298	±26	522	Cohere	CC-BY-NC-4.0
170	134◄─►201	reka-flash-20240904	1295	±27	413	Reka AI	Proprietary
171	139◄─►201	mistral-small-24b-instruct-2501	1294	±24	653	Mistral	Apache 2.0
172	153◄─►193	claude-3-haiku-20240307	1293	±10	6,400	Anthropic	Proprietary
173	139◄─►202	jamba-1.5-large	1292	±27	475	AI21 Labs	Jamba Open
174	142◄─►201	glm-4-0520	1291	±24	558	Zhipu AI	Proprietary
175	134◄─►204	hunyuan-large-vision	1290	±31	390	Tencent	Proprietary
176	128◄─►208	llama-3.1-nemotron-51b-instruct	1290	±39	213	Nvidia	Llama 3.1
177	155◄─►198	qwen2-72b-instruct	1289	±14	2,074	Alibaba	Qianwen LICENSE
178	155◄─►199	gpt-4-0314	1288	±14	2,915	OpenAI	Proprietary
179	159◄─►201	gpt-4-0613	1283	±12	4,778	OpenAI	Proprietary
180	155◄─►202	phi-4	1282	±19	1,037	Microsoft	MIT
181	157◄─►202	amazon-nova-lite-v1.0	1280	±20	873	Amazon	Proprietary
182	159◄─►201	gemini-1.5-flash-001	1280	±12	3,473	Google	Proprietary
183	159◄─►204	amazon-nova-micro-v1.0	1276	±19	953	Amazon	Proprietary
184	160◄─►202	gemini-1.5-flash-8b-001	1275	±14	1,813	Google	Proprietary
185	157◄─►210	jamba-1.5-mini	1274	±27	486	AI21 Labs	Jamba Open
186	159◄─►206	gemini-pro-dev-api	1274	±21	1,078	Google	Proprietary
187	166◄─►204	gemma-2-9b-it	1270	±12	2,823	Google	Gemma license
188	160◄─►208	deepseek-coder-v2	1269	±20	849	DeepSeek AI	DeepSeek License
189	159◄─►218	c4ai-aya-expanse-8b	1268	±27	470	Cohere	CC-BY-NC-4.0
190	168◄─►206	mistral-large-2402	1264	±13	3,439	Mistral	Proprietary
191	157◄─►227	qwen2.5-coder-32b-instruct	1263	±36	221	Alibaba	Apache 2.0
192	162◄─►218	reka-flash-21b-20240226-online	1263	±21	821	Reka AI	Proprietary
193	158◄─►227	ministral-8b-2410	1263	±35	280	Mistral	MRL
194	167◄─►214	reka-flash-21b-20240226	1261	±18	1,302	Reka AI	Proprietary
195	170◄─►211	command-r	1261	±14	2,906	Cohere	CC-BY-NC-4.0
196	168◄─►215	qwen1.5-110b-chat	1259	±17	1,433	Alibaba	Qianwen LICENSE
197	172◄─►214	qwen1.5-72b-chat	1258	±14	2,389	Alibaba	Qianwen LICENSE
198	172◄─►218	mistral-medium	1256	±16	1,927	Mistral	Proprietary
199	157◄─►233	olmo-2-0325-32b-instruct	1255	±46	204	Allen AI	Apache-2.0
200	177◄─►218	mixtral-8x22b-instruct-v0.1	1253	±13	2,829	Mistral	Apache 2.0
201	160◄─►232	gemini-pro	1251	±39	262	Google	Proprietary
202	182◄─►218	llama-3-8b-instruct	1248	±11	5,409	Meta	Llama 3 Community
203	182◄─►222	gpt-3.5-turbo-0125	1247	±12	3,847	OpenAI	Proprietary
204	167◄─►234	zephyr-orpo-141b-A35b-v0.1	1243	±35	288	HuggingFace	Apache 2.0
205	189◄─►227	llama-3.1-8b-instruct	1236	±13	2,583	Meta	Llama 3.1 Community
206	185◄─►229	yi-1.5-34b-chat	1236	±17	1,333	01 AI	Apache-2.0
207	189◄─►230	phi-3-medium-4k-instruct	1232	±17	1,322	Microsoft	MIT
208	191◄─►233	qwen1.5-32b-chat	1227	±18	1,196	Alibaba	Qianwen LICENSE
209	187◄─►236	starling-lm-7b-alpha	1226	±25	568	UC Berkeley	CC-BY-NC-4.0
210	191◄─►235	gemma-1.1-7b-it	1225	±18	1,259	Google	Gemma license
211	185◄─►238	deepseek-llm-67b-chat	1225	±29	385	DeepSeek AI	DeepSeek License
212	199◄─►232	mixtral-8x7b-instruct-v0.1	1224	±12	4,058	Mistral	Apache 2.0
213	193◄─►236	qwen1.5-14b-chat	1222	±20	942	Alibaba	Qianwen LICENSE
214	200◄─►235	gemma-2-2b-it	1221	±13	2,357	Google	Gemma license
215	187◄─►240	openchat-3.5	1221	±30	426	OpenChat	Apache-2.0
216	199◄─►236	dbrx-instruct-preview	1218	±17	1,708	Databricks	DBRX LICENSE
217	193◄─►240	internlm2_5-20b-chat	1217	±26	550	InternLM	Other
218	191◄─►244	llama-3.2-3b-instruct	1216	±31	443	Meta	Llama 3.2
219	199◄─►240	starling-lm-7b-beta	1214	±22	856	Nexusflow	Apache-2.0
220	200◄─►239	phi-3-small-8k-instruct	1214	±19	977	Microsoft	MIT
221	200◄─►240	yi-34b-chat	1212	±22	824	01 AI	Yi License
222	194◄─►247	wizardlm-70b	1212	±30	388	Microsoft	Llama 2 Community
223	190◄─►247	granite-3.0-8b-instruct	1212	±36	356	IBM	Apache 2.0
224	200◄─►240	snowflake-arctic-instruct	1210	±18	1,631	Snowflake	Apache 2.0
225	200◄─►241	openchat-3.5-0106	1209	±21	743	OpenChat	Apache-2.0
226	194◄─►248	openhermes-2.5-mistral-7b	1207	±35	285	NousResearch	Apache-2.0
227	203◄─►247	gpt-3.5-turbo-1106	1200	±22	906	OpenAI	Proprietary
228	199◄─►254	solar-10.7b-instruct-v1.0	1196	±41	210	Upstage AI	CC-BY-NC-4.0
229	209◄─►247	llama-2-70b-chat	1193	±15	2,176	Meta	Llama 2 Community
230	207◄─►247	mistral-7b-instruct-v0.2	1192	±18	1,093	Mistral	Apache-2.0
231	209◄─►248	vicuna-33b	1189	±19	1,205	LMSYS	Non-commercial
232	203◄─►254	wizardlm-13b	1188	±35	326	Microsoft	Llama 2 Community
233	204◄─►254	tulu-2-dpo-70b	1184	±31	348	AllenAI/UW	AI2 ImpACT Low-risk
234	205◄─►257	qwen1.5-7b-chat	1179	±36	267	Alibaba	Qianwen LICENSE
235	215◄─►253	llama-2-13b-chat	1178	±21	976	Meta	Llama 2 Community
236	205◄─►257	qwen-14b-chat	1176	±37	267	Alibaba	Qianwen LICENSE
237	215◄─►254	vicuna-13b	1175	±22	886	LMSYS	Llama 2 Community
238	216◄─►255	gemma-7b-it	1171	±26	538	Google	Gemma license
239	212◄─►257	palm-2	1169	±37	307	Google	Proprietary
240	222◄─►257	zephyr-7b-beta	1164	±26	579	HuggingFace	MIT
241	217◄─►257	nous-hermes-2-mixtral-8x7b-dpo	1163	±32	266	NousResearch	Apache-2.0
242	223◄─►257	phi-3-mini-4k-instruct-june-2024	1161	±26	654	Microsoft	MIT
243	224◄─►257	mistral-7b-instruct	1157	±28	477	Mistral	Apache 2.0
244	224◄─►257	gemma-1.1-2b-it	1156	±27	629	Google	Gemma license
245	223◄─►257	codellama-34b-instruct	1155	±31	352	Meta	Llama 2 Community
246	223◄─►257	vicuna-7b	1151	±35	292	LMSYS	Llama 2 Community
247	224◄─►258	granite-3.0-2b-instruct	1147	±37	380	IBM	Apache 2.0
248	231◄─►257	phi-3-mini-128k-instruct	1145	±22	1,118	Microsoft	MIT
249	231◄─►257	phi-3-mini-4k-instruct	1144	±20	1,081	Microsoft	MIT
250	229◄─►257	stripedhyena-nous-7b	1142	±31	327	Together AI	Apache 2.0
251	231◄─►257	llama-2-7b-chat	1138	±23	766	Meta	Llama 2 Community
252	231◄─►258	olmo-7b-instruct	1134	±32	341	Allen AI	Apache-2.0
253	232◄─►258	qwen1.5-4b-chat	1125	±30	436	Alibaba	Qianwen LICENSE
254	231◄─►259	RWKV-4-Raven-14B	1117	±49	177	RWKV	Apache 2.0
255	237◄─►259	llama-3.2-1b-instruct	1110	±32	446	Meta	Llama 3.2
256	236◄─►259	koala-13b	1107	±42	253	UC Berkeley	Non-commercial
257	237◄─►259	gemma-2b-it	1105	±39	292	Google	Gemma license
258	251◄─►261	alpaca-13b	1061	±50	212	Stanford	Non-commercial
259	254◄─►261	chatglm3-6b	1052	±42	252	Tsinghua	Apache-2.0
260	258◄─►261	oasst-pythia-12b	1002	±49	201	OpenAssistant	Apache 2.0
261	258◄─►261	chatglm-6b	959	±53	193	Tsinghua	Non-commercial

Text Arena

Default Leaderboard Plots

Battle Count for Each Combination of Models (without Ties)

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

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

Confidence Intervals on Model Strength (via Bootstrapping)