LMArena Search (style control)
Arena human-vote leaderboard: rating with confidence interval, vote count and rank. One row per model and category.
No filters. Showing 42 of 42 rows.
1[1–2] OpenAIDossier | xhigh | No harness | GPT-5.6 Sol, Arena rating: 1,255 (scale 1000.0 to 1255.4), 95% confidence interval 1248.8 to 1262.11,255(1248.8–1262.1) | 62364 seeds | Cost kind unknown |
2[1–3] OpenAIDossierTied #1 | max | No harness | GPT-6.1 Sol, Arena rating: 1,248 (scale 1000.0 to 1255.4), 95% confidence interval 1240.5 to 1256.41,248(1240.5–1256.4) | 26556 seeds | Cost kind unknown |
3[1–3] OpenAIDossier | max | No harness | GPT-6 Astra, Arena rating: 1,245 (scale 1000.0 to 1255.4), 95% confidence interval 1236.7 to 1252.81,245(1236.7–1252.8) | 17456 seeds | Cost kind unknown |
4[4–9] OpenAIDossier | max | No harness | GPT-6 Sol, Arena rating: 1,229 (scale 1000.0 to 1255.4), 95% confidence interval 1221.2 to 1236.81,229(1221.2–1236.8) | 13499 seeds | Cost kind unknown |
5[4–9] AnthropicDossier | high | No harness | Fable 5, Arena rating: 1,229 (scale 1000.0 to 1255.4), 95% confidence interval 1221.8 to 1235.31,229(1221.8–1235.3) | 69528 seeds | Cost kind unknown |
6[4–10] AlibabaDossier | No effort set | No harness | Qwen3.8-Max, Arena rating: 1,227 (scale 1000.0 to 1255.4), 95% confidence interval 1213.6 to 1240.91,227(1213.6–1240.9) | 1998 seeds | Cost kind unknown |
7[4–9] AnthropicDossier | high | No harness | Fable 5.1, Arena rating: 1,227 (scale 1000.0 to 1255.4), 95% confidence interval 1219.8 to 1233.81,227(1219.8–1233.8) | 33769 seeds | Cost kind unknown |
8[4–9] OpenAIDossier | search | No harness | GPT-5.5, Arena rating: 1,226 (scale 1000.0 to 1255.4), 95% confidence interval 1220.7 to 1231.21,226(1220.7–1231.2) | 121489 seeds | Cost kind unknown |
9[6–10] AnthropicDossier | search | No harness | Opus 4.6, Arena rating: 1,222 (scale 1000.0 to 1255.4), 95% confidence interval 1217.0 to 1226.51,222(1217.0–1226.5) | 166732 seeds | Cost kind unknown |
10[9–15] AnthropicDossier | high | No harness | Opus 5, Arena rating: 1,213 (scale 1000.0 to 1255.4), 95% confidence interval 1206.5 to 1220.11,213(1206.5–1220.1) | 34000 seeds | Cost kind unknown |
11[9–15] AnthropicDossier | medium | No harness | Opus 5.5, Arena rating: 1,213 (scale 1000.0 to 1255.4), 95% confidence interval 1206.2 to 1219.81,213(1206.2–1219.8) | 31730 seeds | Cost kind unknown |
12[10–16] OpenAIDossier | max | No harness | GPT-6 Luna, Arena rating: 1,211 (scale 1000.0 to 1255.4), 95% confidence interval 1203.9 to 1218.11,211(1203.9–1218.1) | 46776 seeds | Cost kind unknown |
13[10–16] AnthropicDossier | No effort set | No harness | Opus 4.7, Arena rating: 1,210 (scale 1000.0 to 1255.4), 95% confidence interval 1204.7 to 1214.91,210(1204.7–1214.9) | 121890 seeds | Cost kind unknown |
14[11–16] GoogleDossier | grounding | No harness | Gemini 3.1 Pro Preview, Arena rating: 1,208 (scale 1000.0 to 1255.4), 95% confidence interval 1203.7 to 1213.21,208(1203.7–1213.2) | 171494 seeds | Cost kind unknown |
15[11–17] xAIDossier | No effort set | No harness | Grok 4.20 Multi-Agent Beta (0309), Arena rating: 1,207 (scale 1000.0 to 1255.4), 95% confidence interval 1201.3 to 1212.61,207(1201.3–1212.6) | 121069 seeds | Cost kind unknown |
16[12–18] xAIDossier | No effort set | No harness | Grok 4.5, Arena rating: 1,205 (scale 1000.0 to 1255.4), 95% confidence interval 1199.7 to 1211.21,205(1199.7–1211.2) | 105282 seeds | Cost kind unknown |
17[15–21] AnthropicDossier | search | No harness | Sonnet 4.6, Arena rating: 1,201 (scale 1000.0 to 1255.4), 95% confidence interval 1196.4 to 1206.01,201(1196.4–1206.0) | 167499 seeds | Cost kind unknown |
18[15–23] GoogleDossier | grounding | No harness | Gemini 3 Pro Preview, Arena rating: 1,201 (scale 1000.0 to 1255.4), 95% confidence interval 1195.3 to 1206.51,201(1195.3–1206.5) | 38324 seeds | Cost kind unknown |
19[16–23] xAIDossier | No effort set | No harness | Grok 4.20 Beta 1, Arena rating: 1,200 (scale 1000.0 to 1255.4), 95% confidence interval 1193.8 to 1206.11,200(1193.8–1206.1) | 55902 seeds | Cost kind unknown |
20[17–25] OpenAIDossier | search | No harness | o3, Arena rating: 1,197 (scale 1000.0 to 1255.4), 95% confidence interval 1190.4 to 1204.11,197(1190.4–1204.1) | 20655 seeds | Cost kind unknown |
21[18–26] xAIDossier | search | No harness | Grok 4.1 Fast, Arena rating: 1,196 (scale 1000.0 to 1255.4), 95% confidence interval 1190.9 to 1201.11,196(1190.9–1201.1) | 84455 seeds | Cost kind unknown |
22[18–26] OpenAIDossier | search | No harness | GPT-5.4, Arena rating: 1,196 (scale 1000.0 to 1255.4), 95% confidence interval 1189.8 to 1201.31,196(1189.8–1201.3) | 121580 seeds | Cost kind unknown |
23[19–28] AnthropicDossier | No effort set | No harness | Opus 4.8, Arena rating: 1,194 (scale 1000.0 to 1255.4), 95% confidence interval 1188.2 to 1199.71,194(1188.2–1199.7) | 101197 seeds | Cost kind unknown |
24[19–27] xAIDossier | No effort set | No harness | Grok 4.3, Arena rating: 1,194 (scale 1000.0 to 1255.4), 95% confidence interval 1188.8 to 1198.81,194(1188.8–1198.8) | 122703 seeds | Cost kind unknown |
25[19–30] BaiduDossier | No effort set | No harness | ERNIE 5.1, Arena rating: 1,192 (scale 1000.0 to 1255.4), 95% confidence interval 1182.3 to 1201.51,192(1182.3–1201.5) | 3925 seeds | Cost kind unknown |
26[21–30] AnthropicDossier | search | No harness | Sonnet 5, Arena rating: 1,191 (scale 1000.0 to 1255.4), 95% confidence interval 1184.0 to 1197.11,191(1184.0–1197.1) | 71297 seeds | Cost kind unknown |
27[21–30] OpenAIDossier | search | No harness | GPT-5, Arena rating: 1,189 (scale 1000.0 to 1255.4), 95% confidence interval 1181.4 to 1196.41,189(1181.4–1196.4) | 20742 seeds | Cost kind unknown |
28[24–30] GoogleDossier | grounding | No harness | Gemini 3 Flash Preview, Arena rating: 1,189 (scale 1000.0 to 1255.4), 95% confidence interval 1184.0 to 1193.01,189(1184.0–1193.0) | 189590 seeds | Cost kind unknown |
29[25–30] AnthropicDossier | search | No harness | Opus 4.5, Arena rating: 1,186 (scale 1000.0 to 1255.4), 95% confidence interval 1180.4 to 1192.01,186(1180.4–1192.0) | 63904 seeds | Cost kind unknown |
30[27–30] OpenAIDossier | search | No harness | GPT-5.1, Arena rating: 1,184 (scale 1000.0 to 1255.4), 95% confidence interval 1178.4 to 1189.31,184(1178.4–1189.3) | 61989 seeds | Cost kind unknown |
31 AnthropicDossier | search | No harness | Sonnet 4.5, Arena rating: 1,175 (scale 1000.0 to 1255.4), 95% confidence interval 1169.6 to 1180.21,175(1169.6–1180.2) | 139468 seeds | Cost kind unknown |
32[32–33] AnthropicDossier | search | No harness | Opus 4.1, Arena rating: 1,167 (scale 1000.0 to 1255.4), 95% confidence interval 1161.8 to 1171.91,167(1161.8–1171.9) | 78777 seeds | Cost kind unknown |
33[32–34] xAIDossier | search | No harness | Grok 4 Fast, Arena rating: 1,162 (scale 1000.0 to 1255.4), 95% confidence interval 1157.0 to 1166.91,162(1157.0–1166.9) | 42640 seeds | Cost kind unknown |
34[32–34] OpenAIDossier | search | No harness | GPT-5.2, Arena rating: 1,161 (scale 1000.0 to 1255.4), 95% confidence interval 1155.2 to 1167.01,161(1155.2–1167.0) | 54658 seeds | Cost kind unknown |
35[35–36] AnthropicDossier | search | No harness | Opus 4, Arena rating: 1,152 (scale 1000.0 to 1255.4), 95% confidence interval 1146.0 to 1157.31,152(1146.0–1157.3) | 31275 seeds | Cost kind unknown |
36[35–37] GoogleDossier | grounding | No harness | Gemini 2.5 Pro, Arena rating: 1,145 (scale 1000.0 to 1255.4), 95% confidence interval 1140.1 to 1149.91,145(1140.1–1149.9) | 85505 seeds | Cost kind unknown |
37[36–39] PerplexityDossier | high | No harness | Sonar Pro, Arena rating: 1,142 (scale 1000.0 to 1255.4), 95% confidence interval 1136.4 to 1148.41,142(1136.4–1148.4) | 28684 seeds | Cost kind unknown |
38[37–39] PerplexityDossier | high | No harness | Sonar Reasoning Pro, Arena rating: 1,138 (scale 1000.0 to 1255.4), 95% confidence interval 1132.1 to 1144.11,138(1132.1–1144.1) | 29197 seeds | Cost kind unknown |
39[37–39] OpenAIDossier | search · non-thinking | No harness | GPT-5.2, Arena rating: 1,136 (scale 1000.0 to 1255.4), 95% confidence interval 1130.1 to 1141.91,136(1130.1–1141.9) | 78443 seeds | Cost kind unknown |
40 xAIDossier | search | No harness | Grok 4, Arena rating: 1,123 (scale 1000.0 to 1255.4), 95% confidence interval 1116.8 to 1129.31,123(1116.8–1129.3) | 19069 seeds | Cost kind unknown |
41 OpenAIDossier | search | No harness | GPT-4o, Arena rating: 1,084 (scale 1000.0 to 1255.4), 95% confidence interval 1072.1 to 1095.11,084(1072.1–1095.1) | 3392 seeds | Cost kind unknown |
42 DiffbotDossier | No effort set | No harness | Diffbot Small XL, Arena rating: 1,062 (scale 1000.0 to 1255.4), 95% confidence interval 1053.0 to 1070.61,062(1053.0–1070.6) | 6339 seeds | Cost kind unknown |
- 1OpenAIDossier
xhigh
Arena rating
GPT-5.6 Sol, Arena rating: 1,255 (scale 1000.0 to 1255.4), 95% confidence interval 1248.8 to 1262.11,255(1248.8–1262.1) Arena rating
GPT-6.1 Sol, Arena rating: 1,248 (scale 1000.0 to 1255.4), 95% confidence interval 1240.5 to 1256.41,248(1240.5–1256.4)- 3OpenAIDossier
max
Arena rating
GPT-6 Astra, Arena rating: 1,245 (scale 1000.0 to 1255.4), 95% confidence interval 1236.7 to 1252.81,245(1236.7–1252.8) - 4OpenAIDossier
max
Arena rating
GPT-6 Sol, Arena rating: 1,229 (scale 1000.0 to 1255.4), 95% confidence interval 1221.2 to 1236.81,229(1221.2–1236.8) - 5AnthropicDossier
high
Arena rating
Fable 5, Arena rating: 1,229 (scale 1000.0 to 1255.4), 95% confidence interval 1221.8 to 1235.31,229(1221.8–1235.3) - 6AlibabaDossier
Arena rating
Qwen3.8-Max, Arena rating: 1,227 (scale 1000.0 to 1255.4), 95% confidence interval 1213.6 to 1240.91,227(1213.6–1240.9) - 7AnthropicDossier
high
Arena rating
Fable 5.1, Arena rating: 1,227 (scale 1000.0 to 1255.4), 95% confidence interval 1219.8 to 1233.81,227(1219.8–1233.8) - 8OpenAIDossier
search
Arena rating
GPT-5.5, Arena rating: 1,226 (scale 1000.0 to 1255.4), 95% confidence interval 1220.7 to 1231.21,226(1220.7–1231.2) - 9AnthropicDossier
search
Arena rating
Opus 4.6, Arena rating: 1,222 (scale 1000.0 to 1255.4), 95% confidence interval 1217.0 to 1226.51,222(1217.0–1226.5) - 10AnthropicDossier
high
Arena rating
Opus 5, Arena rating: 1,213 (scale 1000.0 to 1255.4), 95% confidence interval 1206.5 to 1220.11,213(1206.5–1220.1) - 11AnthropicDossier
medium
Arena rating
Opus 5.5, Arena rating: 1,213 (scale 1000.0 to 1255.4), 95% confidence interval 1206.2 to 1219.81,213(1206.2–1219.8) - 12OpenAIDossier
max
Arena rating
GPT-6 Luna, Arena rating: 1,211 (scale 1000.0 to 1255.4), 95% confidence interval 1203.9 to 1218.11,211(1203.9–1218.1) - 13AnthropicDossier
Arena rating
Opus 4.7, Arena rating: 1,210 (scale 1000.0 to 1255.4), 95% confidence interval 1204.7 to 1214.91,210(1204.7–1214.9) - 14GoogleDossier
grounding
Arena rating
Gemini 3.1 Pro Preview, Arena rating: 1,208 (scale 1000.0 to 1255.4), 95% confidence interval 1203.7 to 1213.21,208(1203.7–1213.2) - 15xAIDossier
Arena rating
Grok 4.20 Multi-Agent Beta (0309), Arena rating: 1,207 (scale 1000.0 to 1255.4), 95% confidence interval 1201.3 to 1212.61,207(1201.3–1212.6) - 16xAIDossier
Arena rating
Grok 4.5, Arena rating: 1,205 (scale 1000.0 to 1255.4), 95% confidence interval 1199.7 to 1211.21,205(1199.7–1211.2) - 17AnthropicDossier
search
Arena rating
Sonnet 4.6, Arena rating: 1,201 (scale 1000.0 to 1255.4), 95% confidence interval 1196.4 to 1206.01,201(1196.4–1206.0) - 18GoogleDossier
grounding
Arena rating
Gemini 3 Pro Preview, Arena rating: 1,201 (scale 1000.0 to 1255.4), 95% confidence interval 1195.3 to 1206.51,201(1195.3–1206.5) - 19xAIDossier
Arena rating
Grok 4.20 Beta 1, Arena rating: 1,200 (scale 1000.0 to 1255.4), 95% confidence interval 1193.8 to 1206.11,200(1193.8–1206.1) - 20OpenAIDossier
search
Arena rating
o3, Arena rating: 1,197 (scale 1000.0 to 1255.4), 95% confidence interval 1190.4 to 1204.11,197(1190.4–1204.1) - 21xAIDossier
search
Arena rating
Grok 4.1 Fast, Arena rating: 1,196 (scale 1000.0 to 1255.4), 95% confidence interval 1190.9 to 1201.11,196(1190.9–1201.1) - 22OpenAIDossier
search
Arena rating
GPT-5.4, Arena rating: 1,196 (scale 1000.0 to 1255.4), 95% confidence interval 1189.8 to 1201.31,196(1189.8–1201.3) - 23AnthropicDossier
Arena rating
Opus 4.8, Arena rating: 1,194 (scale 1000.0 to 1255.4), 95% confidence interval 1188.2 to 1199.71,194(1188.2–1199.7) - 24xAIDossier
Arena rating
Grok 4.3, Arena rating: 1,194 (scale 1000.0 to 1255.4), 95% confidence interval 1188.8 to 1198.81,194(1188.8–1198.8) - 25BaiduDossier
Arena rating
ERNIE 5.1, Arena rating: 1,192 (scale 1000.0 to 1255.4), 95% confidence interval 1182.3 to 1201.51,192(1182.3–1201.5) - 26AnthropicDossier
search
Arena rating
Sonnet 5, Arena rating: 1,191 (scale 1000.0 to 1255.4), 95% confidence interval 1184.0 to 1197.11,191(1184.0–1197.1) - 27OpenAIDossier
search
Arena rating
GPT-5, Arena rating: 1,189 (scale 1000.0 to 1255.4), 95% confidence interval 1181.4 to 1196.41,189(1181.4–1196.4) - 28GoogleDossier
grounding
Arena rating
Gemini 3 Flash Preview, Arena rating: 1,189 (scale 1000.0 to 1255.4), 95% confidence interval 1184.0 to 1193.01,189(1184.0–1193.0) - 29AnthropicDossier
search
Arena rating
Opus 4.5, Arena rating: 1,186 (scale 1000.0 to 1255.4), 95% confidence interval 1180.4 to 1192.01,186(1180.4–1192.0) - 30OpenAIDossier
search
Arena rating
GPT-5.1, Arena rating: 1,184 (scale 1000.0 to 1255.4), 95% confidence interval 1178.4 to 1189.31,184(1178.4–1189.3) - 31AnthropicDossier
search
Arena rating
Sonnet 4.5, Arena rating: 1,175 (scale 1000.0 to 1255.4), 95% confidence interval 1169.6 to 1180.21,175(1169.6–1180.2) - 32AnthropicDossier
search
Arena rating
Opus 4.1, Arena rating: 1,167 (scale 1000.0 to 1255.4), 95% confidence interval 1161.8 to 1171.91,167(1161.8–1171.9) - 33xAIDossier
search
Arena rating
Grok 4 Fast, Arena rating: 1,162 (scale 1000.0 to 1255.4), 95% confidence interval 1157.0 to 1166.91,162(1157.0–1166.9) - 34OpenAIDossier
search
Arena rating
GPT-5.2, Arena rating: 1,161 (scale 1000.0 to 1255.4), 95% confidence interval 1155.2 to 1167.01,161(1155.2–1167.0) - 35AnthropicDossier
search
Arena rating
Opus 4, Arena rating: 1,152 (scale 1000.0 to 1255.4), 95% confidence interval 1146.0 to 1157.31,152(1146.0–1157.3) - 36GoogleDossier
grounding
Arena rating
Gemini 2.5 Pro, Arena rating: 1,145 (scale 1000.0 to 1255.4), 95% confidence interval 1140.1 to 1149.91,145(1140.1–1149.9) - 37PerplexityDossier
high
Arena rating
Sonar Pro, Arena rating: 1,142 (scale 1000.0 to 1255.4), 95% confidence interval 1136.4 to 1148.41,142(1136.4–1148.4) - 38PerplexityDossier
high
Arena rating
Sonar Reasoning Pro, Arena rating: 1,138 (scale 1000.0 to 1255.4), 95% confidence interval 1132.1 to 1144.11,138(1132.1–1144.1) - 39OpenAIDossier
search · non-thinking
Arena rating
GPT-5.2, Arena rating: 1,136 (scale 1000.0 to 1255.4), 95% confidence interval 1130.1 to 1141.91,136(1130.1–1141.9) - 40xAIDossier
search
Arena rating
Grok 4, Arena rating: 1,123 (scale 1000.0 to 1255.4), 95% confidence interval 1116.8 to 1129.31,123(1116.8–1129.3) - 41OpenAIDossier
search
Arena rating
GPT-4o, Arena rating: 1,084 (scale 1000.0 to 1255.4), 95% confidence interval 1072.1 to 1095.11,084(1072.1–1095.1) - 42DiffbotDossier
Arena rating
Diffbot Small XL, Arena rating: 1,062 (scale 1000.0 to 1255.4), 95% confidence interval 1053.0 to 1070.61,062(1053.0–1070.6)