LMArena Vision: English leaderboard
- models measured
- 85
- lead over #2
- +1 Elo
- median
- 1263
Leader on LMArena Vision: English · read October 22
On October 22, 2026, 95 of our catalog's models had a LMArena Vision: English result, and Anthropic Claude Opus 4.6 led with 1324, ahead of Anthropic Claude Fable 5 at 1323.
Image prompts written in English. People ask a question, compare two anonymous answers and vote for the better one; the rating is built from those votes, as LMArena publishes it.
Role in the index
LMArena Vision: English is shown as published and does not feed any of our indexes. Read the methodology
| # | Model | Result | Votes | Interval (±) | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|---|---|
| 1 | Claude Opus 4.6Anthropic | 1324 | 9,235 | ±8.7 | 57.2 | $25.00 | |
| 2 | Claude Fable 5Anthropic | 1323 | 5,874 | ±9.7 | 79.3 | $50.00 | |
| 3 | Claude Opus 4.7Anthropic | 1322 | 9,834 | ±8.7 | 67.5 | $25.00 | |
| 4 | Gemini 3.8 FlashGoogle | 1318 | 1,556 | ±16.0 | 61.6 | $3.75 | |
| 5 | Claude Opus 5.5Anthropic | 1317 | 992 | ±19.9 | 88.8 | $20.00 | |
| 6 | Claude Opus 5Anthropic | 1315 | 6,237 | ±9.3 | 83.0 | $25.00 | |
| 7 | Claude Fable 5.1Anthropic | 1313 | 1,825 | ±14.6 | 84.2 | $50.00 | |
| 8 | Qwen3.8 Max (0902)Qwen | 1311 | 3,284 | ±11.4 | 62.4 | $6.00 | |
| 9 | Gemini 3.7 FlashGoogle | 1309 | 1,848 | ±14.7 | 58.0 | $3.75 | |
| 10 | Gemini 3.5 FlashGoogle | 1305 | 5,040 | ±9.9 | 39.8 | $9.00 | |
| 11 | GPT-5.4OpenAI | 1301 | 12,132 | ±7.9 | 64.4 | $15.00 | |
| 12 | Claude Sonnet 4.6Anthropic | 1298 | 11,555 | ±8.3 | 57.3 | $15.00 | |
| 13 | Muse Spark 1.3Meta | 1298 | 1,877 | ±14.7 | 57.4 | $4.25 | |
| 14 | GLM 5.3 FlashZ.ai | 1295 | 5,043 | ±9.8 | 57.7 | $0.50 | |
| 15 | Gemini 3.6 FlashGoogle | 1295 | 3,098 | ±11.7 | 62.5 | $3.75 | |
| 16 | Claude Opus 4.8Anthropic | 1293 | 7,311 | ±9.1 | 65.0 | $25.00 | |
| 17 | GPT-5.5OpenAI | 1292 | 11,422 | ±8.0 | 71.3 | $30.00 | |
| 18 | Muse Spark 1.1Meta | 1292 | 4,266 | ±10.4 | 70.3 | $4.25 | |
| 19 | Muse Spark 1.2Meta | 1291 | 695 | ±23.3 | 55.7 | $4.25 | |
| 20 | Step 5 PreviewStepFun | 1290 | 513 | ±26.2 | 61.1 | $2.70 | |
| 21 | Claude Sonnet 5.5Anthropic | 1289 | 859 | ±20.8 | 87.3 | $10.00 | |
| 22 | DeepSeek V4.1 FlashDeepSeek | 1288 | 1,301 | ±17.3 | 62.5 | $0.84 | |
| 23 | Gemini 3.1 Pro PreviewGoogle | 1287 | 18,497 | ±7.0 | 32.3 | $12.00 | |
| 24 | Kimi K2.6Moonshot AI | 1287 | 6,756 | ±9.4 | 52.1 | $4.00 | |
| 25 | Grok 4.5xAI | 1285 | 4,285 | ±10.5 | 58.6 | $6.00 | |
| 26 | Mistral Large 4Mistral AI | 1283 | 355 | ±31.5 | 56.4 | $4.18 | |
| 27 | Qwen3.7 PlusQwen | 1280 | 5,067 | ±9.8 | 33.2 | $1.28 | |
| 28 | GPT-5.6 SolOpenAI | 1280 | 3,912 | ±10.8 | 73.5 | $20.00 | |
| 29 | GPT-6 AstraOpenAI | 1280 | 1,343 | ±16.9 | 85.1 | $50.00 | |
| 30 | Claude Sonnet 5Anthropic | 1275 | 5,335 | ±9.7 | 57.7 | $10.00 | |
| 31 | GLM 5V TurboZ.ai | 1274 | 17,081 | ±7.3 | Not benchmarked | $4.00 | |
| 32 | Kimi K2.5Moonshot AI | 1274 | 10,972 | ±7.7 | 38.6 | $2.85 | |
| 33 | GPT-6.1 SolOpenAI | 1274 | 774 | ±21.7 | 82.4 | $10.00 | |
| 34 | GPT-5.6 TerraOpenAI | 1272 | 4,052 | ±10.7 | 63.9 | $12.00 | |
| 35 | Gemma 4 31BGoogle | 1271 | 15,807 | ±7.7 | 39.0 | $0.40 | |
| 36 | Grok 4.6xAI | 1270 | 2,512 | ±12.9 | 65.7 | $6.00 | |
| 37 | MiMo-V2.6-FlashXiaomi | 1269 | 1,134 | ±18.5 | 54.7 | $0.28 | |
| 38 | Gemini 3.5 Flash LiteGoogle | 1269 | 3,159 | ±11.7 | 49.0 | $2.50 | |
| 39 | Qwen3.5 397B A17BQwen | 1268 | 12,659 | ±7.5 | Not benchmarked | $2.34 | |
| 40 | MiMo-V2.6-ProXiaomi | 1268 | 612 | ±24.3 | 65.4 | $0.87 | |
| 41 | Qwen3.8 27BQwen | 1267 | 2,445 | ±12.9 | 74.2 | $2.55 | |
| 42 | MiniMax M3MiniMax | 1263 | 7,372 | ±9.0 | 39.2 | $1.20 | |
| 43 | GPT-5.6 LunaOpenAI | 1263 | 3,794 | ±10.9 | 58.0 | $1.20 | |
| 44 | Gemini 2.5 ProGoogle | 1261 | 40,672 | ±6.3 | 19.1 | $10.00 | |
| 45 | MiMo-V2.5Xiaomi | 1259 | 11,479 | ±8.1 | 50.0 | $0.336 | |
| 46 | Gemma 4 26B A4BGoogle | 1256 | 9,635 | ±8.5 | Not benchmarked | $0.60 | |
| 47 | Qwen3 VL 235B A22B InstructQwen | 1256 | 4,605 | ±9.8 | Not benchmarked | $1.90 | |
| 48 | GPT-5.1OpenAI | 1252 | 3,532 | ±11.2 | 40.4 | $10.00 | |
| 49 | Qwen3.5-122B-A10BQwen | 1249 | 5,296 | ±9.7 | Not benchmarked | $2.08 | |
| 50 | Qwen3.5-27BQwen | 1246 | 11,800 | ±7.7 | Not benchmarked | $1.56 | |
| 51 | GPT-5.4 MiniOpenAI | 1244 | 10,694 | ±8.3 | 35.8 | $4.50 | |
| 52 | GPT-5.2OpenAI | 1243 | 7,347 | ±8.7 | 55.9 | $14.00 | |
| 53 | T Inkling SmallThinking Machines | 1243 | 2,816 | ±12.2 | 27.3 | $1.20 | |
| 54 | Mistral Medium 3.5Mistral AI | 1240 | 2,314 | ±13.7 | 25.5 | $7.50 | |
| 55 | Gemini 2.5 FlashGoogle | 1239 | 25,447 | ±6.3 | Not benchmarked | $2.50 | |
| 56 | Grok 4.3xAI | 1235 | 11,160 | ±8.1 | 37.4 | $2.50 | |
| 57 | Gemini 3.1 Flash Lite PreviewGoogle | 1232 | 11,005 | ±8.2 | 22.7 | $1.50 | |
| 58 | GPT-5OpenAI | 1231 | 18,691 | ±9.7 | 38.8 | $10.00 | |
| 59 | GPT-6 SolOpenAI | 1229 | 1,058 | ±19.1 | 76.0 | $10.00 | |
| 60 | Grok 4.7xAI | 1229 | 1,292 | ±17.6 | 66.7 | $6.00 | |
| 61 | o3OpenAI | 1221 | 21,609 | ±8.6 | Not benchmarked | $8.00 | |
| 62 | GPT-5 MiniOpenAI | 1216 | 13,428 | ±10.5 | 34.2 | $2.00 | |
| 63 | GPT-4.1OpenAI | 1209 | 19,011 | ±9.0 | Not benchmarked | $8.00 | |
| 64 | GPT-6 LunaOpenAI | 1204 | 1,264 | ±18.1 | 57.3 | $0.50 | |
| 65 | GPT-5.4 NanoOpenAI | 1202 | 10,341 | ±8.5 | 44.8 | $1.25 | |
| 66 | o4 MiniOpenAI | 1200 | 19,813 | ±9.2 | Not benchmarked | $4.40 | |
| 67 | GPT-4.1 MiniOpenAI | 1191 | 19,159 | ±9.5 | Not benchmarked | $1.60 | |
| 68 | Gemma 3 27BGoogle | 1188 | 7,938 | ±10.3 | Not benchmarked | $0.30 | |
| 69 | Claude Opus 4RetiredAnthropic | 1186 | 954 | ±18.5 | Not benchmarked | $75.00 | |
| 70 | GLM 4.5VZ.ai | 1183 | 1,596 | ±15.7 | Not benchmarked | $1.80 | |
| 71 | Mistral Medium 3.1Mistral AI | 1179 | 19,018 | ±8.4 | Not benchmarked | $2.00 | |
| 72 | GLM 4.6VZ.ai | 1179 | 914 | ±19.9 | Not benchmarked | $0.90 | |
| 73 | Gemini 2.0 FlashRetiredGoogle | 1174 | 4,961 | ±9.4 | Not benchmarked | $0.40 | |
| 74 | Mistral Medium 3Mistral AI | 1172 | 5,311 | ±10.6 | Not benchmarked | $2.00 | |
| 75 | GPT-5 NanoOpenAI | 1170 | 1,783 | ±15.2 | Not benchmarked | $0.40 | |
| 76 | o1OpenAI | 1162 | 1,943 | ±14.5 | Not benchmarked | $60.00 | |
| 77 | Gemini 1.5 ProRetiredGoogle | 1159 | 4,829 | ±11.7 | Not benchmarked | $5.00 | |
| 78 | Mistral Small 3.1 24BMistral AI | 1157 | 14,751 | ±10.4 | Not benchmarked | $0.555 | |
| 79 | Gemini 1.5 FlashRetiredGoogle | 1157 | 3,888 | ±12.6 | Not benchmarked | $0.30 | |
| 80 | Llama 4 MaverickMeta Llama | 1150 | 3,387 | ±11.7 | Not benchmarked | $1.00 | |
| 81 | GPT-4o (2024-05-13)OpenAI | 1147 | 14,077 | ±11.3 | Not benchmarked | $15.00 | |
| 82 | Llama 4 ScoutMeta Llama | 1143 | 3,098 | ±12.4 | Not benchmarked | $0.50 | |
| 83 | Claude 3.7 SonnetRetiredAnthropic | 1141 | 2,156 | ±12.9 | Not benchmarked | $15.00 | |
| 84 | Claude 3.5 SonnetRetiredAnthropic | 1124 | 5,352 | ±10.5 | Not benchmarked | $15.00 | |
| 85 | Qwen2.5 VL 72B InstructQwen | 1115 | 2,169 | ±13.8 | Not benchmarked | $1.00 | |
| 86 | GPT-4.1 NanoOpenAI | 1101 | 461 | ±26.5 | Not benchmarked | $0.40 | |
| 87 | GPT-4 TurboOpenAI | 1097 | 8,417 | ±13.8 | Not benchmarked | $30.00 | |
| 88 | Claude 3.5 HaikuRetiredAnthropic | 1088 | 652 | ±23.2 | Not benchmarked | $4.00 | |
| 89 | GPT-4o-mini (2024-07-18)OpenAI | 1072 | 10,721 | ±10.8 | Not benchmarked | $0.60 | |
| 90 | GPT-4o (2024-08-06)OpenAI | 1072 | 1,744 | ±16.5 | Not benchmarked | $10.00 | |
| 91 | Claude 3 OpusRetiredAnthropic | 1018 | 9,732 | ±13.3 | Not benchmarked | $75.00 | |
| 92 | Nova Lite 1.0Amazon | 999 | 1,048 | ±19.7 | Not benchmarked | $0.24 | |
| 93 | Nova Pro 1.0Amazon | 987 | 1,264 | ±18.6 | Not benchmarked | $3.20 | |
| 94 | Claude 3 SonnetRetiredAnthropic | 980 | 7,799 | ±13.9 | Not benchmarked | $15.00 | |
| 95 | Claude 3 HaikuRetiredAnthropic | 949 | 8,449 | ±14.8 | Not benchmarked | $1.25 |
Only models in our catalog are listed, and each appears once with its best published run. 184 rows were read from the source on October 22, 2026, and rows that match no catalog model are left out. Every other column is a figure the board publishes for the same run.