Leaderboard
AI model leaderboard
Which models people prefer when they compare two answers side by side, from LMArena’s open votes. It measures preference, not correctness, and the ± range shows how sure the ranking is. Prices from our API price table.
Questions about images. Leaderboard of 2026-11-09. Checked .
| # | Model | Score | Votes | Licence | API $/M in · out |
|---|---|---|---|---|---|
| 1 | Claude Fable 5 (High) anthropic | 1326 ±7 | 15,536 | Proprietary | — |
| 2 | Claude Opus 5.5 (High) anthropic | 1324 ±10 | 4,789 | Proprietary | — |
| 3 | Claude Fable 5.1 (Max) anthropic | 1322 ±9 | 6,296 | Proprietary | — |
| 4 | Claude Opus 4.7 anthropic | 1319 ±7 | 23,777 | Proprietary | $5 · $25 |
| 5 | Claude Opus 5 (High) anthropic | 1317 ±6 | 19,345 | Proprietary | — |
| 6 | Gemini 3.8 Flash (High) google | 1317 ±8 | 7,624 | Proprietary | — |
| 7 | Claude Opus 4.6 (High) anthropic | 1316 ±6 | 22,116 | Proprietary | — |
| 8 | Claude Opus 4.7 (High) anthropic | 1314 ±7 | 22,609 | Proprietary | — |
| 9 | Gemini 3.7 Flash (High) google | 1313 ±8 | 6,840 | Proprietary | — |
| 10 | Claude Opus 4.6 anthropic | 1312 ±6 | 26,627 | Proprietary | $5 · $25 |
| 11 | Qwen3.8-Max alibaba | 1312 ±7 | 9,842 | Proprietary | $2 · $6 |
| 12 | Muse Spark 1.3 (Max) meta | 1310 ±8 | 7,683 | Proprietary | — |
| 13 | Gemini 3.5 Flash (High) google | 1308 ±7 | 15,325 | Proprietary | — |
| 14 | Muse Spark meta | 1306 ±9 | 5,843 | Proprietary | — |
| 15 | Gemini 3.5 Flash (Medium) google | 1305 ±7 | 15,418 | Proprietary | — |
| 16 | Muse Spark 1.2 (xHigh) meta | 1305 ±14 | 1,930 | Proprietary | — |
| 17 | Gemini 3 Pro google | 1305 ±7 | 13,547 | Proprietary | — |
| 18 | Muse Spark 1.3 (xHigh) meta | 1301 ±8 | 7,522 | Proprietary | — |
| 19 | GPT-5.4 (High) openai | 1301 ±6 | 32,045 | Proprietary | — |
| 20 | Gemini 3.6 Flash (High) google | 1297 ±8 | 10,352 | Proprietary | — |
| 21 | Gemini 3.1 Pro Preview google | 1296 ±5 | 48,242 | Proprietary | $2 · $12 |
| 22 | GPT-5.5 openai | 1296 ±6 | 30,058 | Proprietary | $5 · $30 |
| 23 | Claude Opus 4.8 (High) anthropic | 1295 ±6 | 18,178 | Proprietary | — |
| 24 | GLM-5.3-Flash zai | 1295 ±6 | 31,530 | MIT | $0.15 · $0.5 |
| 25 | GPT-5.5 (High) openai | 1295 ±6 | 28,348 | Proprietary | — |
| 26 | Muse Spark 1.1 meta | 1292 ±7 | 13,413 | Proprietary | — |
| 27 | GPT-5.4 openai | 1292 ±6 | 22,608 | Proprietary | $2.5 · $15 |
| 28 | Claude Opus 4.8 anthropic | 1289 ±6 | 18,692 | Proprietary | $5 · $25 |
| 29 | Grok 4.5 xai | 1288 ±7 | 13,991 | Proprietary | $2 · $6 |
| 30 | GPT-6.1 Sol (Max) openai | 1286 ±10 | 3,987 | Proprietary | — |
| 31 | Gemini 3 Flash google | 1285 ±5 | 38,830 | Proprietary | $0.5 · $3 |
| 32 | Claude Sonnet 5.5 (xHigh) anthropic | 1284 ±10 | 4,257 | Proprietary | — |
| 33 | Claude Sonnet 4.6 anthropic | 1283 ±6 | 27,201 | Proprietary | $3 · $15 |
| 34 | Kimi K2.6 moonshot | 1282 ±7 | 16,265 | Modified MIT | $0.95 · $4 |
| 35 | GPT-5.6 Sol (xHigh) openai | 1282 ±7 | 12,342 | Proprietary | — |
| 36 | Gemma 4 31B google | 1279 ±6 | 41,787 | Apache 2.0 | — |
| 37 | DeepSeek-V4.1-Flash-20260910 (Max) deepseek | 1279 ±10 | 4,247 | MIT | — |
| 38 | GPT-6 Astra (Max) openai | 1279 ±9 | 5,517 | Proprietary | — |
| 39 | Qwen3.7-Plus alibaba | 1277 ±7 | 15,152 | Proprietary | — |
| 40 | Dola Seed 2.0 Pro bytedance | 1274 ±7 | 10,929 | Proprietary | — |
| 41 | Claude Sonnet 5 (High) anthropic | 1272 ±7 | 14,778 | Proprietary | — |
| 42 | Qwen3.8-27B alibaba | 1270 ±8 | 9,632 | Apache 2.0 | — |
| 43 | GPT-5.6 Terra (xHigh) openai | 1269 ±7 | 13,243 | Proprietary | — |
| 44 | Kimi K2.5 (Thinking) moonshot | 1269 ±6 | 28,014 | Modified MIT | — |
| 45 | Gemini 3.5 Flash-Lite google | 1268 ±7 | 10,800 | Proprietary | $0.3 · $2.5 |
| 46 | GPT-5.2 Chat (20260210) openai | 1268 ±6 | 16,361 | Proprietary | — |
| 47 | MiMo-V2.6-Pro xiaomi | 1266 ±15 | 1,826 | MIT | — |
| 48 | GLM-5V-Turbo zai | 1266 ±5 | 43,637 | Proprietary | $1.2 · $4 |
| 49 | Gemini 3 Flash (Minimal) google | 1265 ±5 | 36,907 | Proprietary | — |
| 50 | Grok 4.6 (High) xai | 1264 ±8 | 8,666 | Proprietary | — |
| 51 | Gemini 2.5 Pro google | 1263 ±4 | 88,553 | Proprietary | $1.25 · $10 |
| 52 | Step 5 Preview stepfun | 1263 ±11 | 3,523 | Proprietary | — |
| 53 | Grok 4.20 Beta (Reasoning, 0309) xai | 1263 ±6 | 26,586 | Proprietary | — |
| 54 | Qwen3.5-397B-A17B alibaba | 1262 ±5 | 33,611 | Apache 2.0 | — |
| 55 | Gemma 4 26B A4B google | 1260 ±6 | 23,747 | Apache 2.0 | — |
| 56 | MiMo-V2.6-Flash xiaomi | 1260 ±9 | 6,906 | MIT | — |
| 57 | Grok 4.20 Multi-Agent Beta (0309) xai | 1259 ±6 | 24,357 | Proprietary | — |
| 58 | GPT-5.6 Luna (xHigh) openai | 1256 ±7 | 11,071 | Proprietary | — |
| 59 | Gemini 2.5 Flash Preview (09-2025) google | 1253 ±10 | 4,780 | Proprietary | — |
| 60 | Kimi K2.5 (Instant) moonshot | 1253 ±11 | 4,093 | Modified MIT | — |
| 61 | GPT-5.5 Instant openai | 1251 ±9 | 7,567 | Proprietary | — |
| 62 | MiniMax-M3 minimax | 1249 ±6 | 21,159 | MiniMax Community License | — |
| 63 | GPT-5.1 (High) openai | 1249 ±8 | 9,794 | Proprietary | — |
| 64 | ERNIE-5.0-Preview (1220) baidu | 1248 ±11 | 3,334 | Proprietary | — |
| 65 | MiMo-V2.5 xiaomi | 1248 ±6 | 29,427 | MIT | — |
| 66 | Qwen3-VL-235B-A22B-Instruct alibaba | 1247 ±7 | 12,672 | Apache 2.0 | $0.4 · $1.6 |
| 67 | GPT-5.2 (High) openai | 1245 ±6 | 18,920 | Proprietary | — |
| 68 | GPT-5.4 mini (High) openai | 1244 ±6 | 25,535 | Proprietary | — |
| 69 | Qwen3.5-122B-A10B alibaba | 1244 ±7 | 13,171 | Apache 2.0 | — |
| 70 | Mistral Large 4 mistral | 1244 ±14 | 2,093 | Modified MIT | $0.7 · $2.1 |
| 71 | ChatGPT-4o Latest (20250326) openai | 1243 ±5 | 23,331 | Proprietary | — |
| 72 | Gemini 3.1 Flash-Lite Preview google | 1240 ±6 | 27,443 | Proprietary | $0.25 · $1.5 |
| 73 | Qwen3.5-27B alibaba | 1240 ±5 | 31,308 | Apache 2.0 | — |
| 74 | Inkling-Small thinky | 1236 ±8 | 9,259 | Apache 2.0 | — |
| 75 | Gemini 2.5 Flash google | 1235 ±5 | 59,919 | Proprietary | $0.3 · $2.5 |
| 76 | GPT-5.1 openai | 1235 ±7 | 10,753 | Proprietary | $1.25 · $10 |
| 77 | GPT-6 Sol (Max) openai | 1233 ±10 | 5,018 | Proprietary | — |
| 78 | GPT-5 Chat openai | 1232 ±7 | 39,843 | Proprietary | $1.25 · $10 |
| 79 | Grok 4.7 (xHigh) xai | 1230 ±9 | 6,181 | Proprietary | — |
| 80 | GPT-5.2 openai | 1229 ±6 | 20,001 | Proprietary | $1.75 · $14 |
| 81 | Grok 4.3 xai | 1229 ±6 | 29,224 | Proprietary | $1.25 · $2.5 |
| 82 | MiMo-V2-Omni xiaomi | 1228 ±7 | 10,599 | Proprietary | — |
| 83 | Mistral Medium 3.5 mistral | 1222 ±9 | 5,330 | Modified MIT | $1.5 · $7.5 |
| 84 | Mistral Large 3 mistral | 1222 ±8 | 10,142 | Apache 2.0 | $0.5 · $1.5 |
| 85 | NVIDIA-Nemotron-3.5-Super-VL-120B-A12B (NVFP4) nvidia | 1214 ±9 | 5,532 | OpenMDW-1.1 | — |
| 86 | o3 (2025-04-16) openai | 1214 ±7 | 45,507 | Proprietary | — |
| 87 | Qwen-VL-Max (2025-08-13) alibaba | 1212 ±12 | 3,159 | Proprietary | — |
| 88 | GPT-6 Luna (Max) openai | 1211 ±9 | 6,202 | Proprietary | — |
| 89 | Grok 4 (0709) xai | 1210 ±7 | 30,130 | Proprietary | — |
| 90 | GPT-4.1 (2025-04-14) openai | 1210 ±7 | 40,946 | Proprietary | — |
| 91 | GPT-5 (High) openai | 1208 ±7 | 34,378 | Proprietary | — |
| 92 | Qwen3-VL-235B-A22B-Thinking alibaba | 1206 ±12 | 2,379 | Apache 2.0 | $0.4 · $4 |
| 93 | GPT-5 mini (High) openai | 1202 ±8 | 28,904 | Proprietary | — |
| 94 | Grok 4.1 Fast (Reasoning) xai | 1201 ±7 | 13,295 | Proprietary | — |
| 95 | Gemini 2.5 Flash-Lite Preview (Non-Thinking, 09-2025) google | 1198 ±9 | 4,849 | Proprietary | — |
| 96 | GPT-5.4 nano (High) openai | 1196 ±6 | 25,407 | Proprietary | — |
| 97 | GPT-4.5 Preview (2025-02-27) openai | 1195 ±12 | 2,925 | Proprietary | — |
| 98 | o4-mini (2025-04-16) openai | 1194 ±7 | 41,276 | Proprietary | — |
| 99 | Claude Opus 4 (Thinking 16K, 20250514) anthropic | 1192 ±16 | 1,287 | Proprietary | — |
| 100 | Claude Sonnet 4 (Thinking 32K, 20250514) anthropic | 1191 ±17 | 1,141 | Proprietary | — |
Data: LMArena leaderboard dataset, licensed CC BY 4.0. Prices are matched by model name and may be missing for models without a public API.