• New Chat
  • Leaderboard
  • Take your chats anywhere

    Create an account to save your chat history across your devices.

    • Terms of UsePrivacy Policy
      Start Voting
      OverviewTextWebDevVisionText-to-ImageImage EditSearchText-to-VideoImage-to-VideoCopilot
      Start Voting
      OverviewTextWebDevVisionText-to-ImageImage EditSearchText-to-VideoImage-to-VideoCopilot

      Text Arena

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

      Last Updated

      Oct 16, 2025

      Total Votes

      1,324,542

      Total Models

      258

      /
      /
      1
      Anthropic
      claude-opus-4-1-20250805-thinking-16k
      1484±79,408
      Anthropic
      Proprietary
      1
      Anthropic
      claude-sonnet-4-5-20250929-thinking-32k
      1482±113,115
      Anthropic
      Proprietary
      2
      Anthropic
      claude-sonnet-4-5-20250929
      1461±113,083
      Anthropic
      Proprietary
      3
      Anthropic
      claude-opus-4-1-20250805
      1464±614,945
      Anthropic
      Proprietary
      3
      gemini-2.5-pro
      1462±623,346
      Google
      Proprietary
      3
      Qwen Icon
      qwen3-max-preview
      1454±78,177
      Alibaba
      Proprietary
      5
      gpt-5-high
      1449±710,174
      OpenAI
      Proprietary
      5
      chatgpt-4o-latest-20250326
      1448±615,050
      OpenAI
      Proprietary
      5
      Anthropic
      claude-opus-4-20250514-thinking-16k
      1447±615,158
      Anthropic
      Proprietary
      5
      Qwen Icon
      qwen3-235b-a22b-instruct-2507
      1445±613,107
      Alibaba
      Apache 2.0
      5
      gpt-5-chat
      1445±79,661
      OpenAI
      Proprietary
      5
      Qwen Icon
      qwen3-max-2025-09-23
      1444±103,529
      Alibaba
      Proprietary
      5
      Qwen Icon
      qwen3-vl-235b-a22b-instruct
      1444±113,153
      Alibaba
      Apache 2.0
      5
      glm-4.6
      1443±122,246
      Z.ai
      MIT
      5
      deepseek-v3.2-exp-thinking
      1440±132,126
      DeepSeek AI
      MIT
      5
      deepseek-v3.1-terminus-thinking
      1439±141,702
      DeepSeek AI
      MIT
      6
      o3-2025-04-16
      1444±620,429
      OpenAI
      Proprietary
      7
      deepseek-v3.1-thinking
      1434±95,178
      DeepSeek
      MIT
      7
      MoonshotAI
      kimi-k2-0905-preview
      1434±94,958
      Moonshot
      Modified MIT
      7
      gpt-4.5-preview-2025-02-27
      1434±103,454
      OpenAI
      Proprietary
      10
      grok-4-fast
      1430±113,172
      xAI
      Proprietary
      13
      Anthropic
      claude-opus-4-20250514
      1432±618,073
      Anthropic
      Proprietary
      13
      deepseek-r1-0528
      1430±87,080
      DeepSeek
      MIT
      13
      deepseek-v3.1
      1429±86,852
      DeepSeek
      MIT
      13
      Anthropic
      claude-sonnet-4-20250514-thinking-32k
      1428±614,520
      Anthropic
      Proprietary
      13
      deepseek-v3.2-exp
      1423±122,395
      DeepSeek AI
      MIT
      13
      deepseek-v3.1-terminus
      1421±131,937
      DeepSeek AI
      MIT
      13
      Anthropic
      claude-haiku-4-5-20251001
      1419±171,203
      Anthropic
      Proprietary
      14
      glm-4.5
      1427±79,965
      Z.ai
      MIT
      14
      mistral-medium-2508
      1427±710,845
      Mistral
      Proprietary
      15
      MoonshotAI
      kimi-k2-0711-preview
      1427±612,193
      Moonshot
      Modified MIT
      15
      gpt-4.1-2025-04-14
      1426±616,679
      OpenAI
      Proprietary
      15
      longcat-flash-chat
      1423±85,631
      Meituan
      MIT
      16
      grok-3-preview-02-24
      1421±610,958
      xAI
      Proprietary
      16
      Qwen Icon
      qwen3-next-80b-a3b-instruct
      1421±86,243
      Alibaba
      Apache 2.0
      17
      grok-4-0709
      1420±612,784
      xAI
      Proprietary
      18
      gemini-2.5-flash-preview-09-2025
      1415±103,338
      Google
      Proprietary
      19
      Qwen Icon
      qwen3-235b-a22b-thinking-2507
      1416±103,923
      Alibaba
      Apache 2.0
      22
      Qwen Icon
      qwen3-vl-235b-a22b-thinking
      1412±112,884
      Alibaba
      Apache 2.0
      24
      deepseek-r1
      1413±94,141
      DeepSeek
      MIT
      25
      Qwen Icon
      qwen3-235b-a22b-no-thinking
      1415±616,777
      Alibaba
      Apache 2.0
      26
      o1-2024-12-17
      1413±86,475
      OpenAI
      Proprietary
      30
      Anthropic
      claude-sonnet-4-20250514
      1413±616,567
      Anthropic
      Proprietary
      30
      Anthropic
      claude-3-7-sonnet-20250219-thinking-32k
      1412±614,079
      Anthropic
      Proprietary
      31
      Qwen Icon
      qwen3-coder-480b-a35b-instruct
      1408±710,403
      Alibaba
      Apache 2.0
      32
      Qwen Icon
      qwen3-30b-a3b-instruct-2507
      1407±79,717
      Alibaba
      Apache 2.0
      33
      gpt-5-mini-high
      1406±87,910
      OpenAI
      Proprietary
      35
      o4-mini-2025-04-16
      1404±616,904
      OpenAI
      Proprietary
      35
      mai-1-preview
      1401±86,126
      Microsoft AI
      Proprietary
      36
      Tencent
      hunyuan-t1-20250711
      1395±131,972
      Tencent
      Proprietary
      37
      deepseek-v3-0324
      1401±617,543
      DeepSeek
      MIT
      40
      o3-mini-high
      1396±94,408
      OpenAI
      Proprietary
      44
      gpt-4.1-mini-2025-04-14
      1396±616,518
      OpenAI
      Proprietary
      46
      mistral-medium-2505
      1394±613,942
      Mistral
      Proprietary
      47
      o1-preview
      1392±88,529
      OpenAI
      Proprietary
      49
      Anthropic
      Claude 3.5 Sonnet (10/22)
      1393±427,639
      Anthropic
      Proprietary
      49
      Tencent
      hunyuan-turbos-20250416
      1387±104,019
      Tencent
      Proprietary
      49
      gemini-2.5-flash-lite-preview-09-2025-no-thinking
      1386±103,339
      Google
      Proprietary
      50
      Qwen Icon
      qwen3-235b-a22b
      1388±610,652
      Alibaba
      Apache 2.0
      51
      glm-4.5-air
      1385±79,683
      Z.ai
      MIT
      51
      Qwen Icon
      qwen3-next-80b-a3b-thinking
      1382±85,381
      Alibaba
      Apache 2.0
      52
      Nvidia
      llama-3.1-nemotron-ultra-253b-v1
      1368±21717
      Nvidia
      Nvidia Open Model
      54
      Minimax
      minimax-m1
      1381±613,396
      MiniMax
      Apache 2.0
      55
      Qwen Icon
      qwen2.5-max
      1380±69,905
      Alibaba
      Proprietary
      55
      gemini-2.5-flash-lite-preview-06-17-thinking
      1380±613,444
      Google
      Proprietary
      56
      Stepfun
      step-3
      1372±113,011
      StepFun
      Apache 2.0
      56
      Tencent
      hunyuan-turbos-20250226
      1360±23536
      Tencent
      Proprietary
      57
      grok-3-mini-high
      1374±77,700
      xAI
      Proprietary
      58
      glm-4.5v
      1368±122,375
      Z.ai
      MIT
      58
      Nvidia
      llama-3.3-nemotron-49b-super-v1
      1354±24510
      Nvidia
      Nvidia
      59
      Qwen Icon
      qwen3-32b
      1361±161,214
      Alibaba
      Apache 2.0
      60
      mistral-small-2506
      1369±77,896
      Mistral
      Apache 2.0
      62
      grok-3-mini-beta
      1367±79,830
      xAI
      Proprietary
      62
      AntGroup
      ling-flash-2.0
      1362±122,245
      Ant Group
      MIT
      65
      o3-mini
      1366±520,391
      OpenAI
      Proprietary
      65
      Nvidia
      nvidia-llama-3.3-nemotron-super-49b-v1.5
      1355±151,481
      Nvidia
      Nvidia Open
      65
      amazon-nova-experimental-chat-10-09
      1355±151,435
      Amazon
      Proprietary
      66
      Cohere
      command-a-03-2025
      1361±518,297
      Cohere
      CC-BY-NC-4.0
      66
      gemma-3-27b-it
      1359±615,781
      Google
      Gemma
      66
      gpt-5-nano-high
      1352±103,824
      OpenAI
      Proprietary
      66
      Tencent
      hunyuan-turbo-0110
      1341±23500
      Tencent
      Proprietary
      69
      Anthropic
      Claude 3.5 Sonnet (06/20)
      1356±523,321
      Anthropic
      Proprietary
      69
      o1-mini
      1355±613,953
      OpenAI
      Proprietary
      69
      Qwen Icon
      qwen-plus-0125
      1347±141,485
      Alibaba
      Proprietary
      70
      Qwen Icon
      qwq-32b
      1352±79,013
      Alibaba
      Apache 2.0
      70
      gpt-oss-120b
      1351±79,389
      OpenAI
      Apache 2.0
      71
      AntGroup
      ring-flash-2.0
      1343±122,358
      Ant Group
      MIT
      72
      gemini-2.0-flash-lite-preview-02-05
      1346±76,225
      Google
      Proprietary
      73
      Gemini-1.5-Pro-002
      1347±514,906
      Google
      Proprietary
      73
      deepseek-v3
      1345±85,430
      DeepSeek
      DeepSeek
      76
      Qwen Icon
      qwen3-30b-a3b
      1340±710,919
      Alibaba
      Apache 2.0
      76
      Stepfun
      step-2-16k-exp-202412
      1332±151,226
      StepFun
      Proprietary
      76
      gemma-3-12b-it
      1329±18982
      Google
      Gemma
      79
      01.AI
      yi-lightning
      1336±86,975
      01 AI
      Proprietary
      80
      Anthropic
      claude-3-5-haiku-20241022
      1338±524,036
      Anthropic
      Propretary
      80
      Meta
      llama-3.1-405b-instruct-bf16
      1337±610,702
      Meta
      Llama 3.1 Community
      80
      Tencent
      hunyuan-large-2025-02-10
      1324±19889
      Tencent
      Proprietary
      82
      Meta
      llama-4-maverick-17b-128e-instruct
      1336±616,260
      Meta
      Llama 4
      82
      gpt-4.1-nano-2025-04-14
      1327±141,672
      OpenAI
      Proprietary
      85
      Meta
      llama-3.1-405b-instruct-fp8
      1332±616,287
      Meta
      Llama 3.1 Community
      85
      Stepfun
      step-1o-turbo-202506
      1329±103,889
      StepFun
      Proprietary
      86
      gpt-4o-2024-05-13
      1332±532,551
      OpenAI
      Proprietary
      86
      magistral-medium-2506
      1329±85,740
      Mistral
      Proprietary
      86
      deepseek-v2.5-1210
      1324±131,795
      DeepSeek
      DeepSeek
      86
      glm-4-plus-0111
      1323±141,470
      Zhipu
      Proprietary
      87
      athene-v2-chat
      1328±76,529
      NexusFlow
      NexusFlow
      90
      Meta
      llama-4-scout-17b-16e-instruct
      1325±613,153
      Meta
      Llama
      90
      Qwen Icon
      qwen2.5-plus-1127
      1322±112,678
      Alibaba
      Proprietary
      94
      Anthropic
      claude-3-opus-20240229
      1324±555,322
      Anthropic
      Proprietary
      94
      gpt-oss-20b
      1319±94,789
      OpenAI
      Apache 2.0
      95
      gemini-1.5-pro-001
      1322±622,597
      Google
      Proprietary
      95
      GPT-4o (08/06)
      1321±612,682
      OpenAI
      Proprietary
      95
      gemini-advanced-0514
      1320±714,171
      Google
      Proprietary
      97
      grok-2-2024-08-13
      1321±517,326
      xAI
      Proprietary
      97
      Meta
      llama-3.3-70b-instruct
      1317±517,224
      Meta
      Llama-3.3
      97
      deepseek-v2.5
      1316±76,887
      DeepSeek
      DeepSeek
      97
      mistral-small-3.1-24b-instruct-2503
      1315±613,452
      Mistral
      Apache 2.0
      97
      mistral-large-2407
      1315±612,732
      Mistral
      Mistral Research
      97
      glm-4-plus
      1314±87,090
      Zhipu AI
      Proprietary
      97
      Qwen Icon
      qwen-max-0919
      1313±94,467
      Alibaba
      Qwen
      97
      Tencent
      hunyuan-standard-2025-02-10
      1305±19875
      Tencent
      Proprietary
      101
      gemma-3n-e4b-it
      1312±78,789
      Google
      Gemma
      102
      Qwen Icon
      qwen2.5-72b-instruct
      1312±610,575
      Alibaba
      Qwen
      102
      Nvidia
      llama-3.1-nemotron-70b-instruct
      1305±131,957
      Nvidia
      Llama 3.1
      104
      gpt-4-turbo-2024-04-09
      1311±628,432
      OpenAI
      Proprietary
      105
      athene-70b-0725
      1307±85,467
      NexusFlow
      CC-BY-NC-4.0
      107
      gpt-4-1106-preview
      1308±626,352
      OpenAI
      Proprietary
      107
      mistral-large-2411
      1308±76,984
      Mistral
      MRL
      109
      Qwen Icon
      qwen2.5-coder-32b-instruct
      1298±151,388
      Alibaba
      Apache 2.0
      110
      Tencent
      hunyuan-large-vision
      1299±132,223
      Tencent
      Proprietary
      112
      GPT-4o-mini (07/18)
      1306±518,367
      OpenAI
      Proprietary
      118
      gpt-4-0125-preview
      1302±625,710
      OpenAI
      Proprietary
      118
      gpt-4-0314
      1299±814,007
      OpenAI
      Proprietary
      119
      Gemini-1.5-Flash-002
      1299±79,292
      Google
      Proprietary
      124
      grok-2-mini-2024-08-13
      1298±614,246
      xAI
      Proprietary
      124
      amazon-nova-pro-v1.0
      1296±76,342
      Amazon
      Proprietary
      127
      Meta
      Meta-Llama-3.1-70B-Instruct
      1294±615,042
      Meta
      Llama 3.1 Community
      128
      gemma-3-4b-it
      1281±171,091
      Google
      Gemma
      129
      Nvidia
      llama-3.1-nemotron-51b-instruct
      1277±171,003
      Nvidia
      Llama 3.1
      133
      deepseek-coder-v2
      1282±104,449
      DeepSeek AI
      DeepSeek License
      133
      jamba-1.5-large
      1278±122,379
      AI21 Labs
      Jamba Open
      134
      gemini-1.5-flash-001
      1283±618,303
      Google
      Proprietary
      134
      gpt-4-0613
      1282±623,133
      OpenAI
      Proprietary
      134
      mistral-small-24b-instruct-2501
      1279±103,584
      Mistral
      Apache 2.0
      135
      reka-core-20240904
      1275±112,159
      Reka AI
      Proprietary
      135
      llama-3.1-tulu-3-70b
      1269±19782
      Ai2
      Llama 3.1
      137
      Nvidia
      nemotron-4-340b-instruct
      1276±95,602
      Nvidia
      NVIDIA Open Model
      137
      glm-4-0520
      1273±112,984
      Zhipu AI
      Proprietary
      137
      Tencent
      hunyuan-standard-256k
      1265±21703
      Tencent
      Proprietary
      138
      Anthropic
      claude-3-sonnet-20240229
      1277±630,897
      Anthropic
      Proprietary
      138
      gemma-2-27b-it
      1276±520,564
      Google
      Gemma license
      138
      Meta
      llama-3-70b-instruct
      1275±546,197
      Meta
      Llama 3 Community
      138
      Azure
      phi-4
      1272±85,767
      Microsoft
      MIT
      138
      gemma-2-9b-it-simpo
      1267±112,564
      Princeton
      MIT
      138
      olmo-2-0325-32b-instruct
      1258±20795
      Allen AI
      Apache-2.0
      140
      Qwen Icon
      qwen2-72b-instruct
      1268±710,817
      Alibaba
      Qianwen LICENSE
      140
      amazon-nova-lite-v1.0
      1266±84,953
      Amazon
      Proprietary
      140
      Cohere
      c4ai-aya-expanse-32b
      1265±77,252
      Cohere
      CC-BY-NC-4.0
      140
      reka-flash-20240904
      1262±112,297
      Reka AI
      Proprietary
      147
      Anthropic
      claude-3-haiku-20240307
      1260±533,845
      Anthropic
      Proprietary
      147
      Cohere
      command-r-plus-08-2024
      1256±102,829
      Cohere
      CC-BY-NC-4.0
      149
      mistral-large-2402
      1257±717,296
      Mistral
      Proprietary
      151
      gemini-1.5-flash-8b-001
      1255±79,694
      Google
      Proprietary
      151
      Cohere
      command-r-08-2024
      1251±102,994
      Cohere
      CC-BY-NC-4.0
      151
      ministral-8b-2410
      1246±151,262
      Mistral
      MRL
      155
      gemma-2-9b-it
      1251±614,813
      Google
      Gemma license
      157
      Cohere
      command-r-plus
      1247±622,502
      Cohere
      CC-BY-NC-4.0
      159
      Qwen Icon
      qwen1.5-110b-chat
      1241±97,746
      Alibaba
      Qianwen LICENSE
      161
      amazon-nova-micro-v1.0
      1240±84,897
      Amazon
      Proprietary
      161
      granite-3.1-8b-instruct
      1227±20801
      IBM
      Apache 2.0
      163
      mixtral-8x22b-instruct-v0.1
      1238±714,802
      Mistral
      Apache 2.0
      164
      Qwen Icon
      qwen1.5-72b-chat
      1234±810,895
      Alibaba
      Qianwen LICENSE
      164
      jamba-1.5-mini
      1231±122,352
      AI21 Labs
      Jamba Open
      165
      reka-flash-21b-20240226-online
      1230±114,542
      Reka AI
      Proprietary
      166
      mistral-medium
      1231±98,694
      Mistral
      Proprietary
      166
      gemini-pro
      1222±181,409
      Google
      Proprietary
      166
      llama-3.1-tulu-3-8b
      1219±20729
      Ai2
      Llama 3.1
      167
      reka-flash-21b-20240226
      1229±97,490
      Reka AI
      Proprietary
      167
      Cohere
      c4ai-aya-expanse-8b
      1225±112,588
      Cohere
      CC-BY-NC-4.0
      167
      gemini-pro-dev-api
      1224±114,530
      Google
      Proprietary
      167
      InternLM
      internlm2_5-20b-chat
      1221±112,605
      InternLM
      Other
      167
      gpt-3.5-turbo-1106
      1221±133,858
      OpenAI
      Proprietary
      167
      granite-3.1-2b-instruct
      1214±19835
      IBM
      Apache 2.0
      171
      gpt-3.5-turbo-0125
      1222±718,653
      OpenAI
      Proprietary
      171
      01.AI
      yi-1.5-34b-chat
      1221±86,712
      01 AI
      Apache-2.0
      171
      HuggingFace
      zephyr-orpo-141b-A35b-v0.1
      1211±171,248
      HuggingFace
      Apache 2.0
      172
      Meta
      llama-3.1-8b-instruct
      1218±613,590
      Meta
      Llama 3.1 Community
      172
      Qwen Icon
      qwen1.5-32b-chat
      1216±96,108
      Alibaba
      Qianwen LICENSE
      173
      Meta
      llama-3-8b-instruct
      1214±630,360
      Meta
      Llama 3 Community
      173
      dbrx-instruct-preview
      1212±98,907
      Databricks
      DBRX LICENSE
      176
      Cohere
      command-r
      1209±715,189
      Cohere
      CC-BY-NC-4.0
      176
      Azure
      phi-3-medium-4k-instruct
      1208±86,839
      Microsoft
      MIT
      180
      mixtral-8x7b-instruct-v0.1
      1206±719,635
      Mistral
      Apache 2.0
      181
      granite-3.0-8b-instruct
      1197±151,669
      IBM
      Apache 2.0
      186
      Qwen Icon
      qwen1.5-14b-chat
      1196±114,940
      Alibaba
      Qianwen LICENSE
      189
      starling-lm-7b-beta
      1191±114,546
      Nexusflow
      Apache-2.0
      189
      tulu-2-dpo-70b
      1183±161,422
      AllenAI/UW
      AI2 ImpACT Low-risk
      194
      gemma-1.1-7b-it
      1185±87,041
      Google
      Gemma license
      194
      Azure
      phi-3-small-8k-instruct
      1183±95,253
      Microsoft
      MIT
      194
      OpenChat
      openchat-3.5-0106
      1182±113,443
      OpenChat
      Apache-2.0
      194
      Snowflake
      snowflake-arctic-instruct
      1181±99,644
      Snowflake
      Apache 2.0
      194
      gemma-2-2b-it
      1180±612,355
      Google
      Gemma license
      194
      01.AI
      yi-34b-chat
      1179±103,850
      01 AI
      Yi License
      194
      openhermes-2.5-mistral-7b
      1175±171,140
      NousResearch
      Apache-2.0
      194
      granite-3.0-2b-instruct
      1171±151,743
      IBM
      Apache 2.0
      194
      deepseek-llm-67b-chat
      1170±191,122
      DeepSeek AI
      DeepSeek License
      194
      OpenChat
      openchat-3.5
      1170±161,664
      OpenChat
      Apache-2.0
      194
      Qwen Icon
      qwq-32b-preview
      1167±19855
      Alibaba
      Apache 2.0
      195
      Azure
      wizardlm-70b
      1168±151,762
      Microsoft
      Llama 2 Community
      195
      Meta
      codellama-70b-instruct
      1152±31308
      Meta
      Llama 2 Community
      196
      Azure
      phi-3-mini-4k-instruct-june-2024
      1170±113,172
      Microsoft
      MIT
      197
      Meta
      llama-3.2-3b-instruct
      1164±122,225
      Meta
      Llama 3.2
      197
      mpt-30b-chat
      1151±27436
      MosaicML
      CC-BY-NC-SA-4.0
      197
      dolphin-2.2.1-mistral-7b
      1150±28359
      Cognitive Computations
      Apache-2.0
      198
      starling-lm-7b-alpha
      1164±132,472
      UC Berkeley
      CC-BY-NC-4.0
      199
      HuggingFace
      smollm2-1.7b-instruct
      1149±25577
      HuggingFace
      Apache 2.0
      200
      solar-10.7b-instruct-v1.0
      1154±20935
      Upstage AI
      CC-BY-NC-4.0
      203
      vicuna-33b
      1159±105,083
      LMSYS
      Non-commercial
      204
      Meta
      llama-2-70b-chat
      1155±89,692
      Meta
      Llama 2 Community
      204
      mistral-7b-instruct-v0.2
      1152±105,175
      Mistral
      Apache-2.0
      204
      gemma-7b-it
      1146±142,279
      Google
      Gemma license
      204
      Qwen Icon
      qwen1.5-7b-chat
      1146±161,335
      Alibaba
      Qianwen LICENSE
      204
      Nvidia
      llama2-70b-steerlm-chat
      1141±21790
      Nvidia
      Llama 2 Community
      207
      Azure
      phi-3-mini-4k-instruct
      1147±95,949
      Microsoft
      MIT
      207
      Qwen Icon
      qwen-14b-chat
      1136±191,008
      Alibaba
      Qianwen LICENSE
      207
      falcon-180b-chat
      1120±36231
      TII
      Falcon-180B TII License
      208
      palm-2
      1138±161,607
      Google
      Proprietary
      213
      HuggingFace
      zephyr-7b-alpha
      1118±30334
      HuggingFace
      MIT
      215
      Meta
      llama-2-13b-chat
      1134±104,498
      Meta
      Llama 2 Community
      215
      nous-hermes-2-mixtral-8x7b-dpo
      1125±181,066
      NousResearch
      Apache-2.0
      216
      gemma-1.1-2b-it
      1129±123,196
      Google
      Gemma license
      216
      Meta
      codellama-34b-instruct
      1127±151,541
      Meta
      Llama 2 Community
      217
      vicuna-13b
      1125±113,983
      LMSYS
      Llama 2 Community
      217
      Azure
      phi-3-mini-128k-instruct
      1125±116,202
      Microsoft
      MIT
      220
      HuggingFace
      zephyr-7b-beta
      1112±142,226
      HuggingFace
      MIT
      220
      Azure
      wizardlm-13b
      1109±171,282
      Microsoft
      Llama 2 Community
      222
      mistral-7b-instruct
      1109±151,924
      Mistral
      Apache 2.0
      222
      guanaco-33b
      1098±26516
      UW
      Non-commercial
      223
      Meta
      llama-3.2-1b-instruct
      1110±132,257
      Meta
      Llama 3.2
      225
      gemma-2b-it
      1104±181,196
      Google
      Gemma license
      227
      stripedhyena-nous-7b
      1102±171,237
      Together AI
      Apache 2.0
      229
      vicuna-7b
      1098±171,262
      LMSYS
      Llama 2 Community
      233
      Meta
      llama-2-7b-chat
      1092±113,293
      Meta
      Llama 2 Community
      233
      Qwen Icon
      qwen1.5-4b-chat
      1088±142,079
      Alibaba
      Qianwen LICENSE
      239
      gpt4all-13b-snoozy
      1055±31347
      Nomic AI
      Non-commercial
      241
      olmo-7b-instruct
      1065±171,477
      Allen AI
      Apache-2.0
      243
      chatglm3-6b
      1054±201,001
      Tsinghua
      Apache-2.0
      245
      chatglm2-6b
      1022±27492
      Tsinghua
      Apache-2.0
      246
      koala-13b
      1027±191,253
      UC Berkeley
      Non-commercial
      246
      mpt-7b-chat
      1024±22714
      MosaicML
      CC-BY-NC-SA-4.0
      247
      oasst-pythia-12b
      1013±191,153
      OpenAssistant
      Apache 2.0
      247
      RWKV
      RWKV-4-Raven-14B
      1008±21877
      RWKV
      Apache 2.0
      248
      alpaca-13b
      997±201,058
      Stanford
      Non-commercial
      248
      chatglm-6b
      994±22848
      Tsinghua
      Non-commercial
      253
      dolly-v2-12b
      961±25616
      Databricks
      MIT
      255
      Stability
      stablelm-tuned-alpha-7b
      935±25576
      Stability AI
      CC-BY-NC-SA-4.0
      255
      fastchat-t5-3b
      930±22797
      LMSYS
      Apache 2.0
      256
      Meta
      llama-13b
      905±31430
      Meta
      Non-commercial

      Default Leaderboard Plots

      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

      Battle Count for Each Combination of Models (without Ties)

      Confidence Intervals on Model Strength (via Bootstrapping)