Text Arena | Overall

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

Feb 19, 2026

5,326,341 votes

186 open source models

	Rank Spread
1	1025	glm-5 Z.ai · MIT	1455±8	5,319
2	1030	qwen3.5-397b-a17b Alibaba · Apache 2.0	1450±10	3,602
3	1229	kimi-k2.5-thinking Moonshot · Modified MIT	1449±6	9,823
4	1334	glm-4.7 Z.ai · MIT	1441±6	11,968
5	2040	kimi-k2.5-instant Moonshot · Modified MIT	1437±8	5,857
6	2844	kimi-k2-thinking-turbo Moonshot · Modified MIT	1430±4	34,970
7	3354	glm-4.6 Z.ai · MIT	1425±4	35,233
8	3157	deepseek-v3.2-exp DeepSeek · MIT	1423±6	11,721
9	3057	deepseek-v3.2-exp-thinking DeepSeek · MIT	1423±7	8,979
10	3654	qwen3-235b-a22b-instruct-2507 Alibaba · Apache 2.0	1422±3	70,609
11	3657	deepseek-v3.2-thinking DeepSeek · MIT	1420±5	24,580
12	3659	deepseek-v3.2 DeepSeek · MIT	1419±5	29,555
13	3662	deepseek-r1-0528 DeepSeek · MIT	1419±6	19,273
14	3662	deepseek-v3.1 DeepSeek · MIT	1418±6	15,260
15	3662	kimi-k2-0905-preview Moonshot · Modified MIT	1417±6	11,955
16	3662	deepseek-v3.1-thinking DeepSeek · MIT	1417±7	11,959
17	3762	kimi-k2-0711-preview Moonshot · Modified MIT	1417±5	28,625
18	3566	deepseek-v3.1-terminus DeepSeek · MIT	1416±10	3,756
19	3670	deepseek-v3.1-terminus-thinking DeepSeek · MIT	1416±10	3,547
20	3766	qwen3-vl-235b-a22b-instruct Alibaba · Apache 2.0	1415±6	11,649
21	4062	mistral-large-3 Mistral · Apache 2.0	1414±5	25,814
22	4471	glm-4.5 Z.ai · MIT	1410±5	24,745
23	6077	qwen3-235b-a22b-no-thinking Alibaba · Apache 2.0	1402±4	39,501
24	6079	qwen3-next-80b-a3b-instruct Alibaba · Apache 2.0	1401±5	22,789
25	5487	minimax-m2.5 MiniMax · Modified MIT	1401±9	4,743
26	6084	longcat-flash-chat Meituan · MIT	1400±6	11,523
27	6387	qwen3-235b-a22b-thinking-2507 Alibaba · Apache 2.0	1399±6	9,237
28	6386	deepseek-r1 DeepSeek · MIT	1398±5	18,537
29	6392	qwen3-vl-235b-a22b-thinking Alibaba · Apache 2.0	1395±7	7,958
30	6689	deepseek-v3-0324 DeepSeek · MIT	1394±4	46,636
31	7094	mimo-v2-flash (non-thinking) Xiaomi · MIT	1390±5	18,406
32	6895	step-3.5-flash StepFun · Apache 2.0	1390±7	7,404
33	7195	mimo-v2-flash (thinking) Xiaomi · MIT	1387±6	10,889
34	7495	qwen3-coder-480b-a35b-instruct Alibaba · Apache 2.0	1386±5	26,589
35	7495	minimax-m2.1-preview MiniMax · MIT	1386±5	17,143
36	7697	qwen3-30b-a3b-instruct-2507 Alibaba · Apache 2.0	1383±5	24,055
37	75108	glm-4.6v Z.ai · MIT	1378±11	2,790
38	90104	qwen3-235b-a22b Alibaba · Apache 2.0	1375±5	27,142
39	95108	glm-4.5-air Z.ai · MIT	1371±4	31,293
40	95111	qwen3-next-80b-a3b-thinking Alibaba · Apache 2.0	1369±6	13,826
41	95111	minimax-m1 MiniMax · Apache 2.0	1367±4	36,789
42	95115	glm-4.7-flash Z.ai · MIT	1366±7	9,315
43	99113	gemma-3-27b-it Google · Gemma	1365±4	48,652
44	102127	deepseek-v3 DeepSeek · DeepSeek	1358±5	21,788
45	102134	intellect-3 Prime Intellect · MIT	1356±8	5,307
46	105131	mistral-small-2506 Mistral · Apache 2.0	1356±5	18,327
47	108133	gpt-oss-120b OpenAI · Apache 2.0	1354±4	30,895
48	104135	glm-4.5v Z.ai · MIT	1353±8	4,968
49	110133	command-a-03-2025 Cohere · CC-BY-NC-4.0	1353±3	57,360
50	107148	llama-3.1-nemotron-ultra-253b-v1 Nvidia · Nvidia Open Model	1347±12	2,546
51	110141	minimax-m2 MiniMax · Apache 2.0	1347±8	6,710
52	110141	ling-flash-2.0 Ant Group · MIT	1347±7	7,032
53	110145	qwen3-32b Alibaba · Apache 2.0	1347±9	3,932
54	110143	step-3 StepFun · Apache 2.0	1347±7	6,585
55	112150	gemma-3-12b-it Google · Gemma	1342±10	3,829
56	113152	nvidia-llama-3.3-nemotron-super-49b-v1.5 Nvidia · Nvidia Open	1341±10	3,429
57	122152	qwq-32b Alibaba · Apache 2.0	1336±4	26,094
58	126151	llama-3.1-405b-instruct-bf16 Meta · Llama 3.1 Community	1335±4	41,392
59	128152	llama-3.1-405b-instruct-fp8 Meta · Llama 3.1 Community	1334±4	59,655
60	130160	olmo-3.1-32b-instruct Allen AI · Apache 2.0	1331±6	12,287
61	113183	molmo-2-8b AI2 · Apache 2.0	1328±21	821
62	133165	qwen3-30b-a3b Alibaba · Apache 2.0	1328±5	27,408
63	134165	llama-4-maverick-17b-128e-instruct Meta · Llama 4	1328±4	41,096
64	125174	llama-3.3-nemotron-49b-super-v1 Nvidia · Nvidia	1327±12	2,230
65	135174	deepseek-v2.5-1210 DeepSeek · DeepSeek	1323±8	6,793
66	144171	llama-4-scout-17b-16e-instruct Meta · Llama	1323±5	31,203
67	144176	ring-flash-2.0 Ant Group · MIT	1320±7	7,190
68	149174	llama-3.3-70b-instruct Meta · Llama-3.3	1320±3	55,552
69	146176	gemma-3n-e4b-it Google · Gemma	1319±5	23,320
70	146177	qwen-max-0919 Alibaba · Qwen	1318±6	16,479
71	147181	gpt-oss-20b OpenAI · Apache 2.0	1317±6	10,791
72	149180	nvidia-nemotron-3-nano-30b-a3b-bf16 Nvidia · NVIDIA Open Model	1317±6	15,444
73	153181	athene-v2-chat NexusFlow · NexusFlow	1314±5	24,746
74	153181	mistral-large-2407 Mistral · Mistral Research	1314±4	45,460
75	165186	deepseek-v2.5 DeepSeek · DeepSeek	1307±5	24,574
76	165186	athene-70b-0725 NexusFlow · CC-BY-NC-4.0	1306±6	19,622
77	162186	olmo-3-32b-think Allen AI · Apache 2.0	1306±8	5,878
78	169186	mistral-large-2411 Mistral · MRL	1305±4	28,081
79	171186	mistral-small-3.1-24b-instruct-2503 Mistral · Apache 2.0	1305±4	34,082
80	164193	gemma-3-4b-it Google · Gemma	1303±9	4,177
81	173186	qwen2.5-72b-instruct Alibaba · Qwen	1303±4	39,409
82	173196	llama-3.1-nemotron-70b-instruct Nvidia · Llama 3.1	1299±8	7,136
83	183197	llama-3.1-70b-instruct Meta · Llama 3.1 Community	1294±4	55,234
84	183201	jamba-1.5-large AI21 Labs · Jamba Open	1289±7	8,659
85	183202	ibm-granite-h-small IBM · Apache 2.0	1288±8	5,657
86	185199	gemma-2-27b-it Google · Gemma license	1288±3	75,764
87	183207	llama-3.1-tulu-3-70b Ai2 · Llama 3.1	1287±10	2,846
88	183207	llama-3.1-nemotron-51b-instruct Nvidia · Llama 3.1	1286±10	3,749
89	185207	olmo-3.1-32b-think Allen AI · Apache 2.0	1285±7	8,455
90	188207	gemma-2-9b-it-simpo Princeton · MIT	1279±7	10,069
91	190207	nemotron-4-340b-instruct Nvidia · NVIDIA Open Model	1278±5	19,661
92	190209	command-r-plus-08-2024 Cohere · CC-BY-NC-4.0	1277±7	9,869
93	195207	llama-3-70b-instruct Meta · Llama 3 Community	1276±4	156,880
94	195210	mistral-small-24b-instruct-2501 Mistral · Apache 2.0	1274±6	14,677
95	195216	qwen2.5-coder-32b-instruct Alibaba · Apache 2.0	1271±8	5,430
96	202216	c4ai-aya-expanse-32b Cohere · CC-BY-NC-4.0	1267±5	27,123
97	204216	gemma-2-9b-it Google · Gemma license	1266±4	54,615
98	203217	deepseek-coder-v2 DeepSeek · DeepSeek License	1264±6	15,147
99	206217	command-r-plus Cohere · CC-BY-NC-4.0	1262±4	77,556
100	205217	qwen2-72b-instruct Alibaba · Qianwen LICENSE	1262±5	37,325
101	210218	phi-4 Microsoft · MIT	1256±5	24,126
102	207224	olmo-2-0325-32b-instruct Allen AI · Apache-2.0	1252±11	3,335
103	214223	command-r-08-2024 Cohere · CC-BY-NC-4.0	1250±7	10,141
104	217231	jamba-1.5-mini AI21 Labs · Jamba Open	1239±7	8,854
105	217235	ministral-8b-2410 Mistral · MRL	1237±9	4,780
106	219235	qwen1.5-110b-chat Alibaba · Qianwen LICENSE	1234±6	26,191
107	219235	qwen1.5-72b-chat Alibaba · Qianwen LICENSE	1233±5	39,296
108	221236	mixtral-8x22b-instruct-v0.1 Mistral · Apache 2.0	1230±5	51,417
109	221237	command-r Cohere · CC-BY-NC-4.0	1227±5	54,038
110	226238	llama-3-8b-instruct Meta · Llama 3 Community	1223±4	104,636
111	222239	c4ai-aya-expanse-8b Cohere · CC-BY-NC-4.0	1223±7	9,827
112	222242	llama-3.1-tulu-3-8b Ai2 · Llama 3.1	1221±11	2,895
113	233242	yi-1.5-34b-chat 01 AI · Apache-2.0	1214±5	24,142
114	228244	zephyr-orpo-141b-A35b-v0.1 HuggingFace · Apache 2.0	1213±11	4,653
115	235242	llama-3.1-8b-instruct Meta · Llama 3.1 Community	1212±4	49,605
116	231248	granite-3.1-8b-instruct IBM · Apache 2.0	1209±11	3,092
117	235248	qwen1.5-32b-chat Alibaba · Qianwen LICENSE	1204±6	21,744
118	239249	gemma-2-2b-it Google · Gemma license	1199±4	46,618
119	239250	phi-3-medium-4k-instruct Microsoft · MIT	1198±5	25,055
120	240250	mixtral-8x7b-instruct-v0.1 Mistral · Apache 2.0	1197±4	73,505
121	240255	dbrx-instruct-preview Databricks · DBRX LICENSE	1195±6	32,196
122	240259	internlm2_5-20b-chat InternLM · Other	1191±7	9,902
123	240259	qwen1.5-14b-chat Alibaba · Qianwen LICENSE	1191±7	17,841
124	243265	wizardlm-70b Microsoft · Llama 2 Community	1185±9	8,214
125	242266	deepseek-llm-67b-chat DeepSeek · DeepSeek License	1185±12	4,933
126	246262	yi-34b-chat 01 AI · Yi License	1184±7	15,483
127	246265	openchat-3.5-0106 OpenChat · Apache-2.0	1183±8	12,636
128	246266	openchat-3.5 OpenChat · Apache-2.0	1182±10	7,967
129	246266	granite-3.0-8b-instruct IBM · Apache 2.0	1182±9	6,643
130	247266	gemma-1.1-7b-it Google · Gemma license	1180±6	23,893
131	247266	snowflake-arctic-instruct Snowflake · Apache 2.0	1180±6	32,836
132	246267	granite-3.1-2b-instruct IBM · Apache 2.0	1179±11	3,191
133	247267	tulu-2-dpo-70b AllenAI/UW · AI2 ImpACT Low-risk	1178±10	6,534
134	247270	openhermes-2.5-mistral-7b NousResearch · Apache-2.0	1175±10	5,006
135	249269	vicuna-33b LMSYS · Non-commercial	1173±6	22,479
136	249271	starling-lm-7b-beta Nexusflow · Apache-2.0	1172±7	16,057
137	249270	phi-3-small-8k-instruct Microsoft · MIT	1171±6	17,763
138	250270	llama-2-70b-chat Meta · Llama 2 Community	1171±6	38,491
139	250273	starling-lm-7b-alpha UC Berkeley · CC-BY-NC-4.0	1168±8	10,224
140	252273	llama-3.2-3b-instruct Meta · Llama 3.2	1167±8	7,936
141	250276	nous-hermes-2-mixtral-8x7b-dpo NousResearch · Apache-2.0	1165±12	3,776
142	257282	qwq-32b-preview Alibaba · Apache 2.0	1157±12	3,233
143	263280	granite-3.0-2b-instruct IBM · Apache 2.0	1156±8	6,837
144	259283	llama2-70b-steerlm-chat Nvidia · Llama 2 Community	1156±13	3,584
145	260286	solar-10.7b-instruct-v1.0 Upstage AI · CC-BY-NC-4.0	1153±13	4,155
146	259288	dolphin-2.2.1-mistral-7b Cognitive Computations · Apache-2.0	1152±15	1,679
147	264286	mpt-30b-chat MosaicML · CC-BY-NC-SA-4.0	1150±12	2,571
148	266283	mistral-7b-instruct-v0.2 Mistral · Apache-2.0	1150±7	19,402
149	266285	wizardlm-13b Microsoft · Llama 2 Community	1149±9	7,046
150	264290	falcon-180b-chat TII · Falcon-180B TII License	1147±17	1,295
151	266289	qwen1.5-7b-chat Alibaba · Qianwen LICENSE	1144±10	4,735
152	267288	phi-3-mini-4k-instruct-june-2024 Microsoft · MIT	1143±6	12,296
153	267289	llama-2-13b-chat Meta · Llama 2 Community	1142±7	19,171
154	267289	vicuna-13b LMSYS · Llama 2 Community	1141±7	19,366
155	267291	qwen-14b-chat Alibaba · Qianwen LICENSE	1139±11	4,964
156	268291	codellama-34b-instruct Meta · Llama 2 Community	1137±9	7,363
157	269291	gemma-7b-it Google · Gemma license	1136±10	8,925
158	271292	zephyr-7b-beta HuggingFace · MIT	1131±9	11,116
159	274292	phi-3-mini-128k-instruct Microsoft · MIT	1129±7	20,691
160	276292	phi-3-mini-4k-instruct Microsoft · MIT	1129±6	20,115
161	272296	guanaco-33b UW · Non-commercial	1128±12	2,921
162	271296	zephyr-7b-alpha HuggingFace · MIT	1127±16	1,785
163	279296	stripedhyena-nous-7b Together AI · Apache 2.0	1121±11	5,184
164	274297	codellama-70b-instruct Meta · Llama 2 Community	1119±18	1,143
165	284296	vicuna-7b LMSYS · Llama 2 Community	1115±9	6,923
166	280297	smollm2-1.7b-instruct HuggingFace · Apache 2.0	1114±14	2,201
167	287296	gemma-1.1-2b-it Google · Gemma license	1114±8	10,853
168	287296	llama-3.2-1b-instruct Meta · Llama 3.2	1111±8	8,045
169	287297	mistral-7b-instruct Mistral · Apache 2.0	1110±9	8,977
170	287297	llama-2-7b-chat Meta · Llama 2 Community	1108±7	14,148
171	293301	gemma-2b-it Google · Gemma license	1092±12	4,779
172	297300	qwen1.5-4b-chat Alibaba · Qianwen LICENSE	1090±9	7,598
173	297304	olmo-7b-instruct Allen AI · Apache-2.0	1075±11	6,329
174	298304	koala-13b UC Berkeley · Non-commercial	1071±10	6,964
175	299304	alpaca-13b Stanford · Non-commercial	1068±12	5,745
176	297305	gpt4all-13b-snoozy Nomic AI · Non-commercial	1066±15	1,743
177	299305	mpt-7b-chat MosaicML · CC-BY-NC-SA-4.0	1062±12	3,925
178	299305	chatglm3-6b Tsinghua · Apache-2.0	1056±12	4,658
179	302307	RWKV-4-Raven-14B RWKV · Apache 2.0	1041±11	4,845
180	305307	chatglm2-6b Tsinghua · Apache-2.0	1024±14	2,657
181	305307	oasst-pythia-12b OpenAssistant · Apache 2.0	1022±11	6,311
182	308311	chatglm-6b Tsinghua · Non-commercial	996±13	4,914
183	308311	fastchat-t5-3b LMSYS · Apache 2.0	991±12	4,203
184	308311	dolly-v2-12b Databricks · MIT	980±14	3,412
185	308312	llama-13b Meta · Non-commercial	972±16	2,391
186	311312	stablelm-tuned-alpha-7b Stability AI · CC-BY-NC-SA-4.0	953±13	3,287

Text Arena | Overall

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)