Text Arena

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

Last Updated

Dec 23, 2025

Total Votes

1,180,713

Total Models

291

/

Style Control

	Rank Spread
1	1◄─►2	gemini-3-pro	1490	±10	4,265	Google	Proprietary
2	1◄─►8	gemini-3-flash	1472	±18	1,150	Google	Proprietary
3	2◄─►10	claude-opus-4-5-20251101	1462	±11	3,101	Anthropic	Proprietary
4	2◄─►15	claude-opus-4-5-20251101-thinking-32k	1456	±12	2,889	Anthropic	Proprietary
5	2◄─►13	gemini-2.5-pro	1453	±6	17,436	Google	Proprietary
6	2◄─►17	gemini-3-flash (thinking-minimal)	1452	±18	1,179	Google	Proprietary
7	2◄─►16	claude-sonnet-4-5-20250929	1445	±8	5,837	Anthropic	Proprietary
8	3◄─►16	claude-sonnet-4-5-20250929-thinking-32k	1445	±8	6,756	Anthropic	Proprietary
9	2◄─►16	gpt-4.5-preview-2025-02-27	1444	±10	4,101	OpenAI	Proprietary
10	4◄─►16	claude-opus-4-1-20250805-thinking-16k	1442	±6	10,359	Anthropic	Proprietary
11	5◄─►17	claude-opus-4-1-20250805	1439	±6	13,201	Anthropic	Proprietary
12	4◄─►19	gpt-5.1-high	1439	±10	3,908	OpenAI	Proprietary
13	5◄─►21	grok-4.1-thinking	1435	±10	4,477	xAI	Proprietary
14	6◄─►27	gpt-5.1	1430	±10	4,219	OpenAI	Proprietary
15	4◄─►41	ernie-5.0-preview-1203	1429	±22	679	Baidu	Proprietary
16	12◄─►26	chatgpt-4o-latest-20250326	1427	±5	15,031	OpenAI	Proprietary
17	3◄─►48	gpt-5.2	1426	±25	549	OpenAI	Proprietary
18	10◄─►31	grok-4.1	1425	±10	4,485	xAI	Proprietary
19	12◄─►30	claude-opus-4-20250514-thinking-16k	1424	±7	7,987	Anthropic	Proprietary
20	14◄─►36	claude-opus-4-20250514	1416	±7	9,693	Anthropic	Proprietary
21	14◄─►48	deepseek-v3.2-exp	1413	±12	2,505	DeepSeek AI	MIT
22	14◄─►49	deepseek-v3.2-thinking	1412	±13	2,150	DeepSeek AI	MIT
23	13◄─►54	gpt-5.2-high	1410	±16	1,399	OpenAI	Proprietary
24	15◄─►51	grok-4-1-fast-reasoning	1409	±11	3,010	xAI	Proprietary
25	14◄─►54	ernie-5.0-preview-1103	1408	±15	1,685	Baidu	Proprietary
26	16◄─►52	deepseek-v3.1-thinking	1408	±11	2,820	DeepSeek	MIT
27	17◄─►49	qwen3-max-preview	1408	±8	6,169	Alibaba	Proprietary
28	13◄─►64	deepseek-v3.1-terminus	1407	±21	793	DeepSeek AI	MIT
29	18◄─►51	grok-3-preview-02-24	1405	±7	8,061	xAI	Proprietary
30	16◄─►57	qwen3-max-2025-09-23	1405	±13	1,951	Alibaba	Proprietary
31	18◄─►52	gpt-5-chat	1405	±8	6,716	OpenAI	Proprietary
32	16◄─►58	deepseek-v3.2	1404	±13	2,397	DeepSeek AI	MIT
33	19◄─►52	gemini-2.5-flash	1403	±5	17,451	Google	Proprietary
34	18◄─►54	glm-4.6	1402	±8	6,029	Z.ai	MIT
35	18◄─►56	kimi-k2-thinking-turbo	1402	±10	4,309	Moonshot	Modified MIT
36	19◄─►54	gpt-5-high	1401	±8	7,124	OpenAI	Proprietary
37	14◄─►68	deepseek-v3.1-terminus-thinking	1401	±21	763	DeepSeek AI	MIT
38	19◄─►54	gpt-4.1-2025-04-14	1401	±6	11,566	OpenAI	Proprietary
39	19◄─►63	deepseek-v3.1	1399	±10	3,519	DeepSeek	MIT
40	20◄─►58	grok-4-0709	1398	±7	9,249	xAI	Proprietary
41	20◄─►59	o3-2025-04-16	1397	±6	13,539	OpenAI	Proprietary
42	20◄─►63	claude-sonnet-4-20250514-thinking-32k	1396	±7	7,682	Anthropic	Proprietary
43	20◄─►63	o1-2024-12-17	1396	±7	7,681	OpenAI	Proprietary
44	19◄─►67	deepseek-v3.2-exp-thinking	1395	±13	2,055	DeepSeek AI	MIT
45	20◄─►63	gemini-2.5-flash-preview-09-2025	1395	±8	7,149	Google	Proprietary
46	18◄─►68	grok-4-fast-chat	1394	±15	1,598	xAI	Proprietary
47	22◄─►63	qwen3-235b-a22b-instruct-2507	1394	±6	11,952	Alibaba	Apache 2.0
48	20◄─►67	mistral-large-3	1393	±13	2,244	Mistral	Apache 2.0
49	24◄─►67	deepseek-r1-0528	1390	±9	4,429	DeepSeek	MIT
50	26◄─►67	claude-haiku-4-5-20251001	1390	±8	6,405	Anthropic	Proprietary
51	29◄─►65	claude-3-7-sonnet-20250219-thinking-32k	1390	±7	9,124	Anthropic	Proprietary
52	20◄─►70	qwen3-vl-235b-a22b-instruct	1389	±13	1,967	Alibaba	Apache 2.0
53	34◄─►68	mistral-medium-2508	1386	±6	10,867	Mistral	Proprietary
54	35◄─►69	claude-sonnet-4-20250514	1385	±7	8,913	Anthropic	Proprietary
55	38◄─►69	deepseek-v3-0324	1384	±6	10,565	DeepSeek	MIT
56	36◄─►72	glm-4.5	1383	±8	5,407	Z.ai	MIT
57	34◄─►73	grok-4-fast-reasoning	1383	±9	4,023	xAI	Proprietary
58	39◄─►73	deepseek-r1	1381	±8	5,376	DeepSeek	MIT
59	44◄─►73	o1-preview	1379	±8	8,382	OpenAI	Proprietary
60	39◄─►76	kimi-k2-0905-preview	1379	±12	2,589	Moonshot	Modified MIT
61	45◄─►74	gemini-2.5-flash-lite-preview-06-17-thinking	1377	±7	7,243	Google	Proprietary
62	39◄─►78	qwen3-235b-a22b-thinking-2507	1377	±13	1,954	Alibaba	Apache 2.0
63	46◄─►74	qwen3-235b-a22b-no-thinking	1376	±7	8,302	Alibaba	Apache 2.0
64	46◄─►74	claude-3-7-sonnet-20250219	1376	±6	10,056	Anthropic	Proprietary
65	53◄─►74	claude-3-5-sonnet-20241022	1374	±5	22,125	Anthropic	Proprietary
66	24◄─►97	hunyuan-vision-1.5-thinking	1373	±27	478	Tencent	Proprietary
67	50◄─►78	kimi-k2-0711-preview	1372	±8	6,261	Moonshot	Modified MIT
68	29◄─►96	mimo-v2-flash (non-thinking)	1371	±24	624	Xiaomi	MIT
69	39◄─►86	hunyuan-t1-20250711	1371	±18	1,040	Tencent	Proprietary
70	56◄─►81	gemini-2.5-flash-lite-preview-09-2025-no-thinking	1368	±8	7,238	Google	Proprietary
71	56◄─►81	mistral-medium-2505	1367	±7	7,529	Mistral	Proprietary
72	57◄─►84	qwen3-coder-480b-a35b-instruct	1366	±8	5,154	Alibaba	Apache 2.0
73	55◄─►94	qwen3-vl-235b-a22b-thinking	1363	±14	1,723	Alibaba	Apache 2.0
74	64◄─►86	qwen2.5-max	1361	±7	8,338	Alibaba	Proprietary
75	60◄─►88	mai-1-preview	1361	±10	3,932	Microsoft AI	Proprietary
76	65◄─►89	gpt-4.1-mini-2025-04-14	1358	±7	8,892	OpenAI	Proprietary
77	64◄─►98	hunyuan-turbos-20250416	1357	±11	2,656	Tencent	Proprietary
78	65◄─►101	longcat-flash-chat	1355	±12	2,487	Meituan	MIT
79	67◄─►97	gpt-5-mini-high	1355	±8	6,075	OpenAI	Proprietary
80	67◄─►94	o4-mini-2025-04-16	1355	±6	10,269	OpenAI	Proprietary
81	67◄─►97	deepseek-v3	1354	±8	5,915	DeepSeek	DeepSeek
82	69◄─►95	gemini-2.0-flash-001	1354	±6	10,774	Google	Proprietary
83	69◄─►96	gemini-1.5-pro-002	1353	±6	14,973	Google	Proprietary
84	69◄─►97	gemma-3-27b-it	1353	±6	10,992	Google	Gemma
85	70◄─►103	qwen3-235b-a22b	1348	±8	6,384	Alibaba	Apache 2.0
86	70◄─►106	grok-3-mini-high	1346	±10	3,693	xAI	Proprietary
87	72◄─►104	gemini-2.0-flash-lite-preview-02-05	1346	±7	6,708	Google	Proprietary
88	73◄─►103	command-a-03-2025	1346	±6	13,019	Cohere	CC-BY-NC-4.0
89	72◄─►106	qwen3-next-80b-a3b-instruct	1344	±9	4,933	Alibaba	Apache 2.0
90	76◄─►106	gpt-4o-2024-05-13	1342	±6	29,993	OpenAI	Proprietary
91	74◄─►110	grok-3-mini-beta	1342	±9	5,008	xAI	Proprietary
92	74◄─►110	qwen3-30b-a3b-instruct-2507	1342	±8	5,156	Alibaba	Apache 2.0
93	74◄─►109	glm-4.5-air	1342	±7	6,993	Z.ai	MIT
94	77◄─►115	o3-mini-high	1339	±9	4,771	OpenAI	Proprietary
95	84◄─►111	gpt-4o-2024-08-06	1338	±7	12,159	OpenAI	Proprietary
96	84◄─►117	gemini-advanced-0514	1337	±8	13,239	Google	Proprietary
97	83◄─►121	qwen3-next-80b-a3b-thinking	1335	±11	3,013	Alibaba	Apache 2.0
98	87◄─►119	minimax-m1	1332	±7	8,029	MiniMax	Apache 2.0
99	79◄─►128	glm-4-plus-0111	1332	±15	1,463	Zhipu	Proprietary
100	88◄─►119	gemini-1.5-pro-001	1331	±7	20,975	Google	Proprietary
101	74◄─►145	hunyuan-turbos-20250226	1330	±21	667	Tencent	Proprietary
102	91◄─►121	claude-3-5-sonnet-20240620	1329	±6	21,933	Anthropic	Proprietary
103	84◄─►142	intellect-3	1328	±17	1,266	Prime Intellect	MIT
104	74◄─►148	amazon-nova-experimental-chat-11-10	1328	±22	731	Amazon	Proprietary
105	88◄─►125	mistral-small-2506	1327	±9	3,941	Mistral	Apache 2.0
106	92◄─►121	gpt-4-turbo-2024-04-09	1327	±6	24,581	OpenAI	Proprietary
107	95◄─►122	grok-2-2024-08-13	1325	±6	17,403	xAI	Proprietary
108	96◄─►125	o3-mini	1323	±6	13,681	OpenAI	Proprietary
109	85◄─►151	hunyuan-turbo-0110	1321	±20	715	Tencent	Proprietary
110	91◄─►148	qwen-plus-0125	1320	±15	1,516	Alibaba	Proprietary
111	85◄─►154	llama-3.1-nemotron-ultra-253b-v1	1319	±21	714	Nvidia	Nvidia Open Model
112	94◄─►151	step-2-16k-exp-202412	1318	±15	1,327	StepFun	Proprietary
113	91◄─►152	glm-4.5v	1318	±18	1,073	Z.ai	MIT
114	96◄─►153	step-3	1314	±15	1,417	StepFun	Apache 2.0
115	103◄─►145	claude-3-opus-20240229	1314	±5	50,058	Anthropic	Proprietary
116	102◄─►146	llama-3.1-405b-instruct-fp8	1314	±6	16,014	Meta	Llama 3.1 Community
117	95◄─►162	gemma-3-12b-it	1313	±18	1,014	Google	Gemma
118	97◄─►152	deepseek-v2.5-1210	1313	±13	1,933	DeepSeek	DeepSeek
119	103◄─►148	llama-4-maverick-17b-128e-instruct	1312	±7	8,999	Meta	Llama 4
120	99◄─►152	amazon-nova-experimental-chat-10-20	1312	±12	2,433	Amazon	Proprietary
121	105◄─►148	gpt-4o-mini-2024-07-18	1312	±5	18,596	OpenAI	Proprietary
122	106◄─►148	llama-3.1-405b-instruct-bf16	1310	±6	11,170	Meta	Llama 3.1 Community
123	106◄─►148	claude-3-5-haiku-20241022	1310	±5	17,172	Anthropic	Proprietary
124	105◄─►151	gpt-oss-120b	1310	±8	6,693	OpenAI	Apache 2.0
125	106◄─►151	o1-mini	1308	±6	14,292	OpenAI	Proprietary
126	106◄─►151	gemini-1.5-flash-002	1308	±7	9,456	Google	Proprietary
127	95◄─►170	llama-3.3-nemotron-49b-super-v1	1308	±23	619	Nvidia	Nvidia
128	106◄─►152	qwq-32b	1308	±8	6,309	Alibaba	Apache 2.0
129	97◄─►164	hunyuan-large-2025-02-10	1307	±19	892	Tencent	Proprietary
130	99◄─►163	qwen3-32b	1307	±17	1,123	Alibaba	Apache 2.0
131	106◄─►152	llama-4-scout-17b-16e-instruct	1307	±8	6,905	Meta	Llama
132	95◄─►173	amazon-nova-experimental-chat-10-09	1307	±23	606	Amazon	Proprietary
133	106◄─►153	glm-4-plus	1307	±8	7,188	Zhipu AI	Proprietary
134	106◄─►153	gpt-4-1106-preview	1306	±6	24,721	OpenAI	Proprietary
135	106◄─►162	qwen-max-0919	1305	±9	4,645	Alibaba	Qwen
136	106◄─►161	qwen3-30b-a3b	1304	±8	6,169	Alibaba	Apache 2.0
137	107◄─►158	gpt-4-0125-preview	1303	±6	23,542	OpenAI	Proprietary
138	107◄─►163	yi-lightning	1301	±8	7,638	01 AI	Proprietary
139	106◄─►166	step-1o-turbo-202506	1301	±13	1,955	StepFun	Proprietary
140	105◄─►170	minimax-m2	1301	±16	1,440	MiniMax	Apache 2.0
141	110◄─►163	mistral-large-2407	1300	±7	12,227	Mistral	Mistral Research
142	106◄─►170	nova-2-lite	1300	±14	2,066	Amazon	Proprietary
143	110◄─►163	gemma-3n-e4b-it	1299	±9	5,122	Google	Gemma
144	106◄─►174	gpt-4.1-nano-2025-04-14	1297	±14	1,584	OpenAI	Proprietary
145	106◄─►175	gemma-3-4b-it	1297	±17	1,083	Google	Gemma
146	103◄─►178	nvidia-llama-3.3-nemotron-super-49b-v1.5	1297	±22	692	Nvidia	Nvidia Open
147	116◄─►164	mistral-large-2411	1296	±7	7,429	Mistral	MRL
148	126◄─►164	gemma-2-27b-it	1294	±5	20,411	Google	Gemma license
149	109◄─►175	gpt-5-nano-high	1294	±14	1,728	OpenAI	Proprietary
150	130◄─►170	grok-2-mini-2024-08-13	1292	±6	14,502	xAI	Proprietary
151	107◄─►179	hunyuan-standard-2025-02-10	1291	±18	940	Tencent	Proprietary
152	116◄─►176	magistral-medium-2506	1291	±12	2,490	Mistral	Proprietary
153	131◄─►170	llama-3.3-70b-instruct	1291	±5	14,016	Meta	Llama-3.3
154	130◄─►175	mistral-small-3.1-24b-instruct-2503	1289	±7	7,251	Mistral	Apache 2.0
155	132◄─►175	gemini-1.5-flash-001	1289	±7	16,478	Google	Proprietary
156	129◄─►177	qwen2.5-plus-1127	1288	±11	2,825	Alibaba	Proprietary
157	116◄─►182	hunyuan-large-vision	1286	±18	1,129	Tencent	Proprietary
158	121◄─►181	ring-flash-2.0	1285	±15	1,486	Ant Group	MIT
159	141◄─►178	gpt-4-0613	1282	±7	22,128	OpenAI	Proprietary
160	138◄─►179	deepseek-v2.5	1282	±8	6,624	DeepSeek	DeepSeek
161	141◄─►179	gpt-4-0314	1280	±8	12,881	OpenAI	Proprietary
162	142◄─►178	qwen2.5-72b-instruct	1280	±7	10,847	Alibaba	Qwen
163	130◄─►185	nvidia-nemotron-3-nano-30b-a3b-bf16	1280	±18	1,131	Nvidia	NVIDIA Open Model
164	131◄─►185	olmo-3-32b-think	1279	±17	1,259	Allen AI	Apache 2.0
165	134◄─►185	ling-flash-2.0	1279	±16	1,461	Ant Group	MIT
166	147◄─►180	athene-v2-chat	1277	±7	6,689	NexusFlow	NexusFlow
167	131◄─►187	llama-3.1-tulu-3-70b	1277	±20	757	Ai2	Llama 3.1
168	147◄─►182	athene-70b-0725	1276	±9	5,373	NexusFlow	CC-BY-NC-4.0
169	142◄─►184	command-r-plus-08-2024	1276	±11	2,669	Cohere	CC-BY-NC-4.0
170	142◄─►185	jamba-1.5-large	1274	±13	2,343	AI21 Labs	Jamba Open
171	142◄─►185	gpt-oss-20b	1274	±13	2,227	OpenAI	Apache 2.0
172	152◄─►183	amazon-nova-pro-v1.0	1272	±7	6,628	Amazon	Proprietary
173	149◄─►187	gemma-2-9b-it-simpo	1271	±12	2,790	Princeton	MIT
174	148◄─►187	reka-core-20240904	1270	±13	1,900	Reka AI	Proprietary
175	130◄─►197	mercury	1268	±29	440	Inception AI	Proprietary
176	158◄─►185	claude-3-sonnet-20240229	1267	±6	27,373	Anthropic	Proprietary
177	147◄─►192	llama-3.1-nemotron-51b-instruct	1267	±18	1,008	Nvidia	Llama 3.1
178	161◄─►187	gemma-2-9b-it	1265	±6	14,671	Google	Gemma license
179	154◄─►191	llama-3.1-nemotron-70b-instruct	1264	±14	1,829	Nvidia	Llama 3.1
180	163◄─►187	llama-3.1-70b-instruct	1262	±6	14,971	Meta	Llama 3.1 Community
181	155◄─►195	ibm-granite-h-small	1258	±18	1,179	IBM	Apache 2.0
182	166◄─►190	command-r-plus	1258	±7	19,599	Cohere	CC-BY-NC-4.0
183	165◄─►194	nemotron-4-340b-instruct	1257	±9	5,392	Nvidia	NVIDIA Open Model
184	162◄─►195	reka-flash-20240904	1256	±13	1,953	Reka AI	Proprietary
185	173◄─►194	gemini-1.5-flash-8b-001	1253	±7	9,513	Google	Proprietary
186	173◄─►195	c4ai-aya-expanse-32b	1252	±8	7,312	Cohere	CC-BY-NC-4.0
187	167◄─►196	glm-4-0520	1252	±12	2,611	Zhipu AI	Proprietary
188	178◄─►195	llama-3-70b-instruct	1248	±6	37,795	Meta	Llama 3 Community
189	178◄─►197	mistral-small-24b-instruct-2501	1246	±9	3,948	Mistral	Apache 2.0
190	179◄─►197	claude-3-haiku-20240307	1244	±6	29,953	Anthropic	Proprietary
191	181◄─►198	qwen2-72b-instruct	1240	±8	9,989	Alibaba	Qianwen LICENSE
192	180◄─►201	deepseek-coder-v2	1238	±11	3,891	DeepSeek AI	DeepSeek License
193	183◄─►199	amazon-nova-lite-v1.0	1238	±8	5,268	Amazon	Proprietary
194	181◄─►203	command-r-08-2024	1237	±12	2,600	Cohere	CC-BY-NC-4.0
195	178◄─►213	olmo-2-0325-32b-instruct	1232	±19	876	Allen AI	Apache-2.0
196	191◄─►208	phi-4	1228	±8	6,488	Microsoft	MIT
197	187◄─►214	qwen2.5-coder-32b-instruct	1225	±15	1,538	Alibaba	Apache 2.0
198	192◄─►209	mistral-large-2402	1224	±8	15,235	Mistral	Proprietary
199	188◄─►214	ministral-8b-2410	1223	±16	1,355	Mistral	MRL
200	193◄─►213	gpt-3.5-turbo-0125	1221	±7	16,187	OpenAI	Proprietary
201	193◄─►213	amazon-nova-micro-v1.0	1221	±8	5,267	Amazon	Proprietary
202	194◄─►215	qwen1.5-110b-chat	1217	±9	6,524	Alibaba	Qianwen LICENSE
203	195◄─►216	gemini-pro-dev-api	1213	±12	4,640	Google	Proprietary
204	195◄─►216	qwen1.5-72b-chat	1212	±8	9,509	Alibaba	Qianwen LICENSE
205	195◄─►216	c4ai-aya-expanse-8b	1211	±11	2,748	Cohere	CC-BY-NC-4.0
206	195◄─►218	jamba-1.5-mini	1211	±12	2,353	AI21 Labs	Jamba Open
207	197◄─►216	mixtral-8x22b-instruct-v0.1	1207	±7	12,839	Mistral	Apache 2.0
208	197◄─►219	mistral-medium	1207	±9	8,411	Mistral	Proprietary
209	197◄─►218	command-r	1206	±8	13,542	Cohere	CC-BY-NC-4.0
210	194◄─►224	hunyuan-standard-256k	1206	±21	752	Tencent	Proprietary
211	195◄─►224	gemini-pro	1203	±19	1,318	Google	Proprietary
212	197◄─►221	reka-flash-21b-20240226-online	1202	±12	3,644	Reka AI	Proprietary
213	196◄─►229	llama-3.1-tulu-3-8b	1199	±20	816	Ai2	Llama 3.1
214	203◄─►223	reka-flash-21b-20240226	1196	±10	5,844	Reka AI	Proprietary
215	200◄─►231	zephyr-orpo-141b-A35b-v0.1	1194	±18	1,132	HuggingFace	Apache 2.0
216	202◄─►230	wizardlm-70b	1193	±15	2,314	Microsoft	Llama 2 Community
217	210◄─►224	gemma-2-2b-it	1191	±6	12,553	Google	Gemma license
218	210◄─►225	llama-3-8b-instruct	1190	±7	25,528	Meta	Llama 3 Community
219	206◄─►234	gpt-3.5-turbo-1106	1187	±13	3,934	OpenAI	Proprietary
220	211◄─►232	llama-3.1-8b-instruct	1181	±6	13,526	Meta	Llama 3.1 Community
221	209◄─►244	granite-3.1-8b-instruct	1177	±21	837	IBM	Apache 2.0
222	215◄─►237	dbrx-instruct-preview	1174	±10	7,999	Databricks	DBRX LICENSE
223	216◄─►237	yi-1.5-34b-chat	1174	±9	6,548	01 AI	Apache-2.0
224	211◄─►243	openchat-3.5	1173	±15	1,998	OpenChat	Apache-2.0
225	216◄─►238	phi-3-medium-4k-instruct	1171	±9	6,660	Microsoft	MIT
226	216◄─►238	qwen1.5-32b-chat	1171	±10	5,419	Alibaba	Qianwen LICENSE
227	217◄─►240	vicuna-33b	1169	±10	5,628	LMSYS	Non-commercial
228	218◄─►238	mixtral-8x7b-instruct-v0.1	1169	±7	17,973	Mistral	Apache 2.0
229	207◄─►256	falcon-180b-chat	1168	±32	342	TII	Falcon-180B TII License
230	212◄─►249	deepseek-llm-67b-chat	1167	±19	1,143	DeepSeek AI	DeepSeek License
231	216◄─►250	tulu-2-dpo-70b	1164	±17	1,547	AllenAI/UW	AI2 ImpACT Low-risk
232	220◄─►246	qwen1.5-14b-chat	1163	±11	4,516	Alibaba	Qianwen LICENSE
233	220◄─►247	yi-34b-chat	1162	±11	3,708	01 AI	Yi License
234	221◄─►249	openchat-3.5-0106	1161	±12	3,122	OpenChat	Apache-2.0
235	221◄─►249	snowflake-arctic-instruct	1158	±10	7,260	Snowflake	Apache 2.0
236	219◄─►254	openhermes-2.5-mistral-7b	1156	±18	1,137	NousResearch	Apache-2.0
237	223◄─►253	internlm2_5-20b-chat	1151	±12	2,761	InternLM	Other
238	221◄─►256	wizardlm-13b	1151	±15	1,815	Microsoft	Llama 2 Community
239	227◄─►253	gemma-1.1-7b-it	1149	±9	6,055	Google	Gemma license
240	229◄─►257	vicuna-13b	1143	±11	4,677	LMSYS	Llama 2 Community
241	228◄─►259	granite-3.0-8b-instruct	1142	±15	1,754	IBM	Apache 2.0
242	229◄─►259	llama-3.2-3b-instruct	1140	±14	2,123	Meta	Llama 3.2
243	226◄─►260	nous-hermes-2-mixtral-8x7b-dpo	1140	±20	858	NousResearch	Apache-2.0
244	231◄─►258	phi-3-small-8k-instruct	1139	±10	4,647	Microsoft	MIT
245	230◄─►259	starling-lm-7b-alpha	1138	±13	2,514	UC Berkeley	CC-BY-NC-4.0
246	227◄─►262	solar-10.7b-instruct-v1.0	1137	±21	894	Upstage AI	CC-BY-NC-4.0
247	226◄─►263	mpt-30b-chat	1137	±23	647	MosaicML	CC-BY-NC-SA-4.0
248	227◄─►271	dolphin-2.2.1-mistral-7b	1130	±29	382	Cognitive Computations	Apache-2.0
249	237◄─►260	llama-2-70b-chat	1129	±9	9,600	Meta	Llama 2 Community
250	235◄─►262	starling-lm-7b-beta	1128	±12	4,023	Nexusflow	Apache-2.0
251	231◄─►269	llama2-70b-steerlm-chat	1127	±22	836	Nvidia	Llama 2 Community
252	234◄─►269	granite-3.1-2b-instruct	1127	±21	851	IBM	Apache 2.0
253	235◄─►272	zephyr-7b-alpha	1122	±25	569	HuggingFace	MIT
254	235◄─►272	guanaco-33b	1121	±23	714	UW	Non-commercial
255	240◄─►269	zephyr-7b-beta	1120	±14	2,693	HuggingFace	MIT
256	238◄─►272	qwen-14b-chat	1118	±18	1,232	Alibaba	Qianwen LICENSE
257	238◄─►272	qwq-32b-preview	1116	±21	857	Alibaba	Apache 2.0
258	242◄─►272	vicuna-7b	1113	±15	1,784	LMSYS	Llama 2 Community
259	245◄─►271	mistral-7b-instruct-v0.2	1112	±10	4,711	Mistral	Apache-2.0
260	241◄─►273	qwen1.5-7b-chat	1111	±18	1,094	Alibaba	Qianwen LICENSE
261	247◄─►277	granite-3.0-2b-instruct	1104	±16	1,821	IBM	Apache 2.0
262	250◄─►274	llama-2-13b-chat	1103	±11	4,790	Meta	Llama 2 Community
263	247◄─►277	codellama-34b-instruct	1102	±15	2,005	Meta	Llama 2 Community
264	250◄─►274	phi-3-mini-4k-instruct-june-2024	1101	±11	3,385	Microsoft	MIT
265	249◄─►278	gemma-7b-it	1101	±15	2,197	Google	Gemma license
266	250◄─►278	phi-3-mini-128k-instruct	1097	±12	4,641	Microsoft	MIT
267	250◄─►278	mistral-7b-instruct	1093	±15	2,338	Mistral	Apache 2.0
268	250◄─►278	palm-2	1092	±15	2,159	Google	Proprietary
269	253◄─►278	phi-3-mini-4k-instruct	1092	±10	5,218	Microsoft	MIT
270	250◄─►279	stripedhyena-nous-7b	1090	±18	1,273	Together AI	Apache 2.0
271	255◄─►279	gemma-1.1-2b-it	1087	±13	2,601	Google	Gemma license
272	253◄─►282	smollm2-1.7b-instruct	1078	±27	630	HuggingFace	Apache 2.0
273	265◄─►280	llama-2-7b-chat	1075	±11	3,460	Meta	Llama 2 Community
274	260◄─►282	alpaca-13b	1074	±19	1,219	Stanford	Non-commercial
275	263◄─►282	llama-3.2-1b-instruct	1074	±15	2,055	Meta	Llama 3.2
276	263◄─►282	qwen1.5-4b-chat	1073	±15	1,899	Alibaba	Qianwen LICENSE
277	263◄─►285	gpt4all-13b-snoozy	1057	±32	347	Nomic AI	Non-commercial
278	261◄─►286	codellama-70b-instruct	1055	±37	262	Meta	Llama 2 Community
279	272◄─►284	gemma-2b-it	1053	±19	1,153	Google	Gemma license
280	270◄─►284	mpt-7b-chat	1053	±21	895	MosaicML	CC-BY-NC-SA-4.0
281	273◄─►284	koala-13b	1046	±18	1,516	UC Berkeley	Non-commercial
282	273◄─►286	chatglm3-6b	1039	±20	1,110	Tsinghua	Apache-2.0
283	277◄─►287	olmo-7b-instruct	1023	±18	1,521	Allen AI	Apache-2.0
284	277◄─►289	chatglm2-6b	1013	±22	775	Tsinghua	Apache-2.0
285	280◄─►289	oasst-pythia-12b	1008	±18	1,411	OpenAssistant	Apache 2.0
286	281◄─►289	RWKV-4-Raven-14B	1000	±19	1,099	RWKV	Apache 2.0
287	283◄─►289	fastchat-t5-3b	993	±21	943	LMSYS	Apache 2.0
288	284◄─►289	chatglm-6b	979	±21	1,056	Tsinghua	Non-commercial
289	284◄─►291	dolly-v2-12b	970	±23	732	Databricks	MIT
290	289◄─►291	stablelm-tuned-alpha-7b	928	±24	681	Stability AI	CC-BY-NC-SA-4.0
291	289◄─►291	llama-13b	922	±29	513	Meta	Non-commercial

Text Arena

Default Leaderboard Plots

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)

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