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.
Answers grounded in web search. Leaderboard of 2026-11-09. Checked .
| # | Model | Score | Votes | Licence | API $/M in · out |
|---|---|---|---|---|---|
| 1 | Qwen3.8-Max alibaba | 1260 ±14 | 1,998 | Proprietary | $2 · $6 |
| 2 | GPT-5.6 Sol (xHigh) openai | 1258 ±6 | 62,364 | Proprietary | — |
| 3 | Claude Opus 4.6 (Search) anthropic | 1252 ±5 | 166,732 | Proprietary | — |
| 4 | GPT-5.5 (Search) openai | 1246 ±5 | 121,489 | Proprietary | — |
| 5 | Claude Opus 5 (High) anthropic | 1239 ±7 | 34,000 | Proprietary | — |
| 6 | Claude Fable 5.1 (High) anthropic | 1237 ±7 | 33,769 | Proprietary | — |
| 7 | GPT-6 Astra (Max) openai | 1231 ±8 | 17,456 | Proprietary | — |
| 8 | GPT-6.1 Sol (Max) openai | 1230 ±8 | 26,556 | Proprietary | — |
| 9 | Claude Opus 4.7 anthropic | 1230 ±5 | 121,890 | Proprietary | $5 · $25 |
| 10 | Claude Fable 5 (High) anthropic | 1228 ±7 | 69,528 | Proprietary | — |
| 11 | ERNIE 5.1 baidu | 1227 ±10 | 3,925 | Proprietary | — |
| 12 | Claude Sonnet 4.6 (Search) anthropic | 1221 ±5 | 167,499 | Proprietary | — |
| 13 | Grok 4.5 xai | 1216 ±6 | 105,282 | Proprietary | $2 · $6 |
| 14 | Claude Opus 5.5 (Medium) anthropic | 1213 ±7 | 31,730 | Proprietary | — |
| 15 | Gemini 3.1 Pro (Grounding) google | 1210 ±5 | 171,494 | Proprietary | — |
| 16 | GPT-5.2 (Search) openai | 1207 ±6 | 54,658 | Proprietary | — |
| 17 | Gemini 3 Pro (Grounding) google | 1207 ±6 | 38,324 | Proprietary | — |
| 18 | Grok 4.20 Multi-Agent Beta (0309) xai | 1205 ±6 | 121,069 | Proprietary | — |
| 19 | Claude Opus 4.8 anthropic | 1202 ±6 | 101,197 | Proprietary | $5 · $25 |
| 20 | GPT-5.1 (Search) openai | 1199 ±5 | 61,989 | Proprietary | — |
| 21 | GPT-5.4 (Search) openai | 1198 ±6 | 121,580 | Proprietary | — |
| 22 | Gemini 3 Flash (Grounding) google | 1196 ±4 | 189,590 | Proprietary | — |
| 23 | Grok 4.20 Beta 1 xai | 1191 ±6 | 55,902 | Proprietary | — |
| 24 | Claude Sonnet 5 (Search) anthropic | 1188 ±6 | 71,297 | Proprietary | — |
| 25 | Claude Opus 4.5 (Search) anthropic | 1179 ±6 | 63,904 | Proprietary | — |
| 26 | GPT-6 Sol (Max) openai | 1174 ±7 | 13,499 | Proprietary | — |
| 27 | GPT-5.2 (Search, Non-Reasoning) openai | 1172 ±6 | 78,443 | Proprietary | — |
| 28 | Grok 4.1 Fast (Search) xai | 1172 ±5 | 84,455 | Proprietary | — |
| 29 | Grok 4 Fast (Search) xai | 1171 ±5 | 42,640 | Proprietary | — |
| 30 | Grok 4.3 xai | 1166 ±5 | 122,703 | Proprietary | $1.25 · $2.5 |
| 31 | GPT-6 Luna (Max) openai | 1160 ±7 | 46,776 | Proprietary | — |
| 32 | Claude Sonnet 4.5 (Search) anthropic | 1159 ±5 | 139,468 | Proprietary | — |
| 33 | Claude Opus 4.1 (Search) anthropic | 1148 ±5 | 78,777 | Proprietary | — |
| 34 | o3 (Search) openai | 1145 ±6 | 20,655 | Proprietary | — |
| 35 | Gemini 2.5 Pro (Grounding) google | 1143 ±5 | 85,505 | Proprietary | — |
| 36 | Grok 4 (Search) xai | 1142 ±6 | 19,069 | Proprietary | — |
| 37 | Sonar Reasoning Pro (High) perplexity | 1139 ±6 | 29,197 | Proprietary | — |
| 38 | GPT-5 (Search) openai | 1133 ±6 | 20,742 | Proprietary | — |
| 39 | Sonar Pro (High) perplexity | 1130 ±6 | 28,684 | Proprietary | — |
| 40 | Claude Opus 4 (Search) anthropic | 1127 ±5 | 31,275 | Proprietary | — |
| 41 | Diffbot-Small-XL diffbot | 1022 ±9 | 6,339 | Apache 2.0 | — |
| 42 | GPT-4o (Search) openai | 1006 ±11 | 3,392 | Proprietary | $2.5 · $10 |
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.