LMArena Document (style control)
Arena human-vote leaderboard: rating with confidence interval, vote count and rank. One row per model and category.
No filters. Showing 46 of 46 rows.
1[1–5] AnthropicDossier | high | No harness | Fable 5, Arena rating: 1,500 (scale 1350.0 to 1500.1), 95% confidence interval 1493.4 to 1506.81,500(1493.4–1506.8) | 12357 seeds | Cost kind unknown |
2[1–10] AnthropicDossierTied #1 | max | No harness | Fable 5.1, Arena rating: 1,498 (scale 1350.0 to 1500.1), 95% confidence interval 1483.2 to 1513.61,498(1483.2–1513.6) | 1453 seeds | Cost kind unknown |
3[1–6] AnthropicDossierTied #1 | high | No harness | Opus 4.6, Arena rating: 1,497 (scale 1350.0 to 1500.1), 95% confidence interval 1490.5 to 1503.31,497(1490.5–1503.3) | 29456 seeds | Cost kind unknown |
4[1–6] AnthropicDossierTied #1 | No effort set | No harness | Opus 4.6, Arena rating: 1,496 (scale 1350.0 to 1500.1), 95% confidence interval 1490.3 to 1500.91,496(1490.3–1500.9) | 47264 seeds | Cost kind unknown |
5[2–7] AnthropicDossierTied #1 | No effort set | No harness | Opus 4.7, Arena rating: 1,494 (scale 1350.0 to 1500.1), 95% confidence interval 1488.1 to 1500.81,494(1488.1–1500.8) | 23628 seeds | Cost kind unknown |
6[1–7] AnthropicDossierTied #1 | high | No harness | Opus 4.7, Arena rating: 1,494 (scale 1350.0 to 1500.1), 95% confidence interval 1487.9 to 1500.81,494(1487.9–1500.8) | 23327 seeds | Cost kind unknown |
7[5–11] OpenAIDossier | xhigh | No harness | GPT-5.6 Sol, Arena rating: 1,486 (scale 1350.0 to 1500.1), 95% confidence interval 1479.6 to 1493.41,486(1479.6–1493.4) | 9649 seeds | Cost kind unknown |
8[6–12] AnthropicDossier | high | No harness | Opus 5, Arena rating: 1,486 (scale 1350.0 to 1500.1), 95% confidence interval 1480.2 to 1491.71,486(1480.2–1491.7) | 16189 seeds | Cost kind unknown |
9[6–13] AnthropicDossier | high | No harness | Opus 4.8, Arena rating: 1,485 (scale 1350.0 to 1500.1), 95% confidence interval 1478.5 to 1492.31,485(1478.5–1492.3) | 12069 seeds | Cost kind unknown |
10[7–16] OpenAIDossier | high | No harness | GPT-5.5, Arena rating: 1,481 (scale 1350.0 to 1500.1), 95% confidence interval 1476.0 to 1486.91,481(1476.0–1486.9) | 27324 seeds | Cost kind unknown |
11[7–17] MetaDossier | No effort set | No harness | Muse Spark 1.1, Arena rating: 1,480 (scale 1350.0 to 1500.1), 95% confidence interval 1472.6 to 1487.01,480(1472.6–1487.0) | 8934 seeds | Cost kind unknown |
12[8–17] OpenAIDossier | No effort set | No harness | GPT-5.5, Arena rating: 1,479 (scale 1350.0 to 1500.1), 95% confidence interval 1473.8 to 1484.61,479(1473.8–1484.6) | 27751 seeds | Cost kind unknown |
13[10–19] AnthropicDossier | No effort set | No harness | Sonnet 4.6, Arena rating: 1,478 (scale 1350.0 to 1500.1), 95% confidence interval 1472.2 to 1484.01,478(1472.2–1484.0) | 61055 seeds | Cost kind unknown |
14[9–19] AnthropicDossier | No effort set | No harness | Opus 4.8, Arena rating: 1,476 (scale 1350.0 to 1500.1), 95% confidence interval 1469.7 to 1482.41,476(1469.7–1482.4) | 14729 seeds | Cost kind unknown |
15[11–23] OpenAIDossier | xhigh | No harness | GPT-5.6 Terra, Arena rating: 1,474 (scale 1350.0 to 1500.1), 95% confidence interval 1467.7 to 1480.71,474(1467.7–1480.7) | 11412 seeds | Cost kind unknown |
16[11–23] MetaDossier | max | No harness | Muse Spark 1.3, Arena rating: 1,474 (scale 1350.0 to 1500.1), 95% confidence interval 1466.3 to 1482.01,474(1466.3–1482.0) | 7205 seeds | Cost kind unknown |
17[11–23] OpenAIDossier | max | No harness | GPT-6 Astra, Arena rating: 1,474 (scale 1350.0 to 1500.1), 95% confidence interval 1465.3 to 1482.61,474(1465.3–1482.6) | 5723 seeds | Cost kind unknown |
18[13–23] OpenAIDossier | No effort set | No harness | GPT-5.4, Arena rating: 1,471 (scale 1350.0 to 1500.1), 95% confidence interval 1465.2 to 1477.51,471(1465.2–1477.5) | 35183 seeds | Cost kind unknown |
19[15–26] GoogleDossier | high | No harness | Gemini 3.8 Flash, Arena rating: 1,469 (scale 1350.0 to 1500.1), 95% confidence interval 1461.4 to 1477.41,469(1461.4–1477.4) | 6950 seeds | Cost kind unknown |
20[15–25] AnthropicDossier | high | No harness | Sonnet 5, Arena rating: 1,469 (scale 1350.0 to 1500.1), 95% confidence interval 1462.3 to 1475.51,469(1462.3–1475.5) | 11044 seeds | Cost kind unknown |
21[15–26] GoogleDossier | high | No harness | Gemini 3.7 Flash, Arena rating: 1,468 (scale 1350.0 to 1500.1), 95% confidence interval 1460.3 to 1475.61,468(1460.3–1475.6) | 7680 seeds | Cost kind unknown |
22[13–29] AnthropicDossier | No effort set | No harness | Opus 4.5, Arena rating: 1,467 (scale 1350.0 to 1500.1), 95% confidence interval 1457.3 to 1477.71,467(1457.3–1477.7) | 8304 seeds | Cost kind unknown |
23[9–33] MetaDossier | No effort set | No harness | Muse Spark, Arena rating: 1,467 (scale 1350.0 to 1500.1), 95% confidence interval 1448.5 to 1484.81,467(1448.5–1484.8) | 1116 seeds | Cost kind unknown |
24[16–27] xAIDossier | No effort set | No harness | Grok 4.5, Arena rating: 1,466 (scale 1350.0 to 1500.1), 95% confidence interval 1458.6 to 1473.91,466(1458.6–1473.9) | 7447 seeds | Cost kind unknown |
25[22–31] OpenAIDossier | xhigh | No harness | GPT-5.6 Luna, Arena rating: 1,460 (scale 1350.0 to 1500.1), 95% confidence interval 1453.0 to 1467.41,460(1453.0–1467.4) | 8832 seeds | Cost kind unknown |
26[21–31] OpenAIDossier | No effort set | No harness | GPT-5.5 Instant, Arena rating: 1,460 (scale 1350.0 to 1500.1), 95% confidence interval 1451.8 to 1468.71,460(1451.8–1468.7) | 8834 seeds | Cost kind unknown |
27[23–31] GoogleDossier | No effort set | No harness | Gemini 3.1 Pro Preview, Arena rating: 1,459 (scale 1350.0 to 1500.1), 95% confidence interval 1453.7 to 1464.21,459(1453.7–1464.2) | 52244 seeds | Cost kind unknown |
28[22–33] GoogleDossier | medium | No harness | Gemini 3.5 Flash, Arena rating: 1,459 (scale 1350.0 to 1500.1), 95% confidence interval 1450.9 to 1466.61,459(1450.9–1466.6) | 6835 seeds | Cost kind unknown |
29[23–36] GoogleDossier | high | No harness | Gemini 3.5 Flash, Arena rating: 1,455 (scale 1350.0 to 1500.1), 95% confidence interval 1446.0 to 1464.51,455(1446.0–1464.5) | 4248 seeds | Cost kind unknown |
30[24–36] GoogleDossier | high | No harness | Gemini 3.6 Flash, Arena rating: 1,454 (scale 1350.0 to 1500.1), 95% confidence interval 1443.4 to 1464.91,454(1443.4–1464.9) | 3031 seeds | Cost kind unknown |
31[27–36] GoogleDossier | No effort set | No harness | Gemma 4 31B, Arena rating: 1,452 (scale 1350.0 to 1500.1), 95% confidence interval 1445.2 to 1458.91,452(1445.2–1458.9) | 15431 seeds | Cost kind unknown |
32[26–36] GoogleDossier | No effort set | No harness | Gemini 3 Pro Preview, Arena rating: 1,451 (scale 1350.0 to 1500.1), 95% confidence interval 1442.2 to 1459.71,451(1442.2–1459.7) | 11338 seeds | Cost kind unknown |
33[27–36] Moonshot AIDossier | No effort set | No harness | Kimi K2.6, Arena rating: 1,450 (scale 1350.0 to 1500.1), 95% confidence interval 1442.8 to 1458.01,450(1442.8–1458.0) | 11809 seeds | Cost kind unknown |
34[28–37] xAIDossier | high | No harness | Grok 4.6, Arena rating: 1,450 (scale 1350.0 to 1500.1), 95% confidence interval 1442.6 to 1458.11,450(1442.6–1458.1) | 7669 seeds | Cost kind unknown |
35[30–37] AnthropicDossier | No effort set | No harness | Sonnet 4.5, Arena rating: 1,448 (scale 1350.0 to 1500.1), 95% confidence interval 1441.8 to 1454.11,448(1441.8–1454.1) | 32916 seeds | Cost kind unknown |
36[30–38] AlibabaDossier | No effort set | No harness | Qwen3.7 Plus, Arena rating: 1,446 (scale 1350.0 to 1500.1), 95% confidence interval 1438.9 to 1453.91,446(1438.9–1453.9) | 7426 seeds | Cost kind unknown |
37[34–40] xAIDossier | thinking | No harness | Grok 4.20 0309, Arena rating: 1,440 (scale 1350.0 to 1500.1), 95% confidence interval 1432.9 to 1446.91,440(1432.9–1446.9) | 21926 seeds | Cost kind unknown |
38[34–41] GoogleDossier | No effort set | No harness | Gemini 3 Flash Preview, Arena rating: 1,440 (scale 1350.0 to 1500.1), 95% confidence interval 1430.5 to 1449.01,440(1430.5–1449.0) | 7538 seeds | Cost kind unknown |
39[36–41] Moonshot AIDossier | thinking | No harness | Kimi K2.5, Arena rating: 1,438 (scale 1350.0 to 1500.1), 95% confidence interval 1430.8 to 1444.31,438(1430.8–1444.3) | 22601 seeds | Cost kind unknown |
40[37–42] MiniMaxDossier | No effort set | No harness | MiniMax M3, Arena rating: 1,435 (scale 1350.0 to 1500.1), 95% confidence interval 1426.8 to 1443.11,435(1426.8–1443.1) | 6617 seeds | Cost kind unknown |
41[38–42] GoogleDossier | No effort set | No harness | Gemini 2.5 Pro, Arena rating: 1,433 (scale 1350.0 to 1500.1), 95% confidence interval 1426.4 to 1438.81,433(1426.4–1438.8) | 26635 seeds | Cost kind unknown |
42[41–44] OpenAIDossier | No effort set | No harness | GPT-5.2, Arena rating: 1,425 (scale 1350.0 to 1500.1), 95% confidence interval 1419.3 to 1431.01,425(1419.3–1431.0) | 29726 seeds | Cost kind unknown |
43[41–45] OpenAIDossier | high | No harness | GPT-5.2, Arena rating: 1,422 (scale 1350.0 to 1500.1), 95% confidence interval 1412.5 to 1430.91,422(1412.5–1430.9) | 7552 seeds | Cost kind unknown |
44[42–45] AnthropicDossier | No effort set | No harness | Haiku 4.5, Arena rating: 1,419 (scale 1350.0 to 1500.1), 95% confidence interval 1413.3 to 1424.11,419(1413.3–1424.1) | 39423 seeds | Cost kind unknown |
45[42–46] OpenAIDossier | No effort set | No harness | GPT-5.1, Arena rating: 1,416 (scale 1350.0 to 1500.1), 95% confidence interval 1407.2 to 1425.41,416(1407.2–1425.4) | 8736 seeds | Cost kind unknown |
46[45–46] Z.aiDossier | No effort set | No harness | GLM-5V-Turbo, Arena rating: 1,406 (scale 1350.0 to 1500.1), 95% confidence interval 1398.0 to 1413.11,406(1398.0–1413.1) | 9136 seeds | Cost kind unknown |
- 1AnthropicDossier
high
Arena rating
Fable 5, Arena rating: 1,500 (scale 1350.0 to 1500.1), 95% confidence interval 1493.4 to 1506.81,500(1493.4–1506.8) Arena rating
Fable 5.1, Arena rating: 1,498 (scale 1350.0 to 1500.1), 95% confidence interval 1483.2 to 1513.61,498(1483.2–1513.6)Arena rating
Opus 4.6, Arena rating: 1,497 (scale 1350.0 to 1500.1), 95% confidence interval 1490.5 to 1503.31,497(1490.5–1503.3)Arena rating
Opus 4.6, Arena rating: 1,496 (scale 1350.0 to 1500.1), 95% confidence interval 1490.3 to 1500.91,496(1490.3–1500.9)Arena rating
Opus 4.7, Arena rating: 1,494 (scale 1350.0 to 1500.1), 95% confidence interval 1488.1 to 1500.81,494(1488.1–1500.8)Arena rating
Opus 4.7, Arena rating: 1,494 (scale 1350.0 to 1500.1), 95% confidence interval 1487.9 to 1500.81,494(1487.9–1500.8)- 7OpenAIDossier
xhigh
Arena rating
GPT-5.6 Sol, Arena rating: 1,486 (scale 1350.0 to 1500.1), 95% confidence interval 1479.6 to 1493.41,486(1479.6–1493.4) - 8AnthropicDossier
high
Arena rating
Opus 5, Arena rating: 1,486 (scale 1350.0 to 1500.1), 95% confidence interval 1480.2 to 1491.71,486(1480.2–1491.7) - 9AnthropicDossier
high
Arena rating
Opus 4.8, Arena rating: 1,485 (scale 1350.0 to 1500.1), 95% confidence interval 1478.5 to 1492.31,485(1478.5–1492.3) - 10OpenAIDossier
high
Arena rating
GPT-5.5, Arena rating: 1,481 (scale 1350.0 to 1500.1), 95% confidence interval 1476.0 to 1486.91,481(1476.0–1486.9) - 11MetaDossier
Arena rating
Muse Spark 1.1, Arena rating: 1,480 (scale 1350.0 to 1500.1), 95% confidence interval 1472.6 to 1487.01,480(1472.6–1487.0) - 12OpenAIDossier
Arena rating
GPT-5.5, Arena rating: 1,479 (scale 1350.0 to 1500.1), 95% confidence interval 1473.8 to 1484.61,479(1473.8–1484.6) - 13AnthropicDossier
Arena rating
Sonnet 4.6, Arena rating: 1,478 (scale 1350.0 to 1500.1), 95% confidence interval 1472.2 to 1484.01,478(1472.2–1484.0) - 14AnthropicDossier
Arena rating
Opus 4.8, Arena rating: 1,476 (scale 1350.0 to 1500.1), 95% confidence interval 1469.7 to 1482.41,476(1469.7–1482.4) - 15OpenAIDossier
xhigh
Arena rating
GPT-5.6 Terra, Arena rating: 1,474 (scale 1350.0 to 1500.1), 95% confidence interval 1467.7 to 1480.71,474(1467.7–1480.7) - 16MetaDossier
max
Arena rating
Muse Spark 1.3, Arena rating: 1,474 (scale 1350.0 to 1500.1), 95% confidence interval 1466.3 to 1482.01,474(1466.3–1482.0) - 17OpenAIDossier
max
Arena rating
GPT-6 Astra, Arena rating: 1,474 (scale 1350.0 to 1500.1), 95% confidence interval 1465.3 to 1482.61,474(1465.3–1482.6) - 18OpenAIDossier
Arena rating
GPT-5.4, Arena rating: 1,471 (scale 1350.0 to 1500.1), 95% confidence interval 1465.2 to 1477.51,471(1465.2–1477.5) - 19GoogleDossier
high
Arena rating
Gemini 3.8 Flash, Arena rating: 1,469 (scale 1350.0 to 1500.1), 95% confidence interval 1461.4 to 1477.41,469(1461.4–1477.4) - 20AnthropicDossier
high
Arena rating
Sonnet 5, Arena rating: 1,469 (scale 1350.0 to 1500.1), 95% confidence interval 1462.3 to 1475.51,469(1462.3–1475.5) - 21GoogleDossier
high
Arena rating
Gemini 3.7 Flash, Arena rating: 1,468 (scale 1350.0 to 1500.1), 95% confidence interval 1460.3 to 1475.61,468(1460.3–1475.6) - 22AnthropicDossier
Arena rating
Opus 4.5, Arena rating: 1,467 (scale 1350.0 to 1500.1), 95% confidence interval 1457.3 to 1477.71,467(1457.3–1477.7) - 23MetaDossier
Arena rating
Muse Spark, Arena rating: 1,467 (scale 1350.0 to 1500.1), 95% confidence interval 1448.5 to 1484.81,467(1448.5–1484.8) - 24xAIDossier
Arena rating
Grok 4.5, Arena rating: 1,466 (scale 1350.0 to 1500.1), 95% confidence interval 1458.6 to 1473.91,466(1458.6–1473.9) - 25OpenAIDossier
xhigh
Arena rating
GPT-5.6 Luna, Arena rating: 1,460 (scale 1350.0 to 1500.1), 95% confidence interval 1453.0 to 1467.41,460(1453.0–1467.4) - 26OpenAIDossier
Arena rating
GPT-5.5 Instant, Arena rating: 1,460 (scale 1350.0 to 1500.1), 95% confidence interval 1451.8 to 1468.71,460(1451.8–1468.7) - 27GoogleDossier
Arena rating
Gemini 3.1 Pro Preview, Arena rating: 1,459 (scale 1350.0 to 1500.1), 95% confidence interval 1453.7 to 1464.21,459(1453.7–1464.2) - 28GoogleDossier
medium
Arena rating
Gemini 3.5 Flash, Arena rating: 1,459 (scale 1350.0 to 1500.1), 95% confidence interval 1450.9 to 1466.61,459(1450.9–1466.6) - 29GoogleDossier
high
Arena rating
Gemini 3.5 Flash, Arena rating: 1,455 (scale 1350.0 to 1500.1), 95% confidence interval 1446.0 to 1464.51,455(1446.0–1464.5) - 30GoogleDossier
high
Arena rating
Gemini 3.6 Flash, Arena rating: 1,454 (scale 1350.0 to 1500.1), 95% confidence interval 1443.4 to 1464.91,454(1443.4–1464.9) - 31GoogleDossier
Arena rating
Gemma 4 31B, Arena rating: 1,452 (scale 1350.0 to 1500.1), 95% confidence interval 1445.2 to 1458.91,452(1445.2–1458.9) - 32GoogleDossier
Arena rating
Gemini 3 Pro Preview, Arena rating: 1,451 (scale 1350.0 to 1500.1), 95% confidence interval 1442.2 to 1459.71,451(1442.2–1459.7) - 33Moonshot AIDossier
Arena rating
Kimi K2.6, Arena rating: 1,450 (scale 1350.0 to 1500.1), 95% confidence interval 1442.8 to 1458.01,450(1442.8–1458.0) - 34xAIDossier
high
Arena rating
Grok 4.6, Arena rating: 1,450 (scale 1350.0 to 1500.1), 95% confidence interval 1442.6 to 1458.11,450(1442.6–1458.1) - 35AnthropicDossier
Arena rating
Sonnet 4.5, Arena rating: 1,448 (scale 1350.0 to 1500.1), 95% confidence interval 1441.8 to 1454.11,448(1441.8–1454.1) - 36AlibabaDossier
Arena rating
Qwen3.7 Plus, Arena rating: 1,446 (scale 1350.0 to 1500.1), 95% confidence interval 1438.9 to 1453.91,446(1438.9–1453.9) - 37xAIDossier
thinking
Arena rating
Grok 4.20 0309, Arena rating: 1,440 (scale 1350.0 to 1500.1), 95% confidence interval 1432.9 to 1446.91,440(1432.9–1446.9) - 38GoogleDossier
Arena rating
Gemini 3 Flash Preview, Arena rating: 1,440 (scale 1350.0 to 1500.1), 95% confidence interval 1430.5 to 1449.01,440(1430.5–1449.0) - 39Moonshot AIDossier
thinking
Arena rating
Kimi K2.5, Arena rating: 1,438 (scale 1350.0 to 1500.1), 95% confidence interval 1430.8 to 1444.31,438(1430.8–1444.3) - 40MiniMaxDossier
Arena rating
MiniMax M3, Arena rating: 1,435 (scale 1350.0 to 1500.1), 95% confidence interval 1426.8 to 1443.11,435(1426.8–1443.1) - 41GoogleDossier
Arena rating
Gemini 2.5 Pro, Arena rating: 1,433 (scale 1350.0 to 1500.1), 95% confidence interval 1426.4 to 1438.81,433(1426.4–1438.8) - 42OpenAIDossier
Arena rating
GPT-5.2, Arena rating: 1,425 (scale 1350.0 to 1500.1), 95% confidence interval 1419.3 to 1431.01,425(1419.3–1431.0) - 43OpenAIDossier
high
Arena rating
GPT-5.2, Arena rating: 1,422 (scale 1350.0 to 1500.1), 95% confidence interval 1412.5 to 1430.91,422(1412.5–1430.9) - 44AnthropicDossier
Arena rating
Haiku 4.5, Arena rating: 1,419 (scale 1350.0 to 1500.1), 95% confidence interval 1413.3 to 1424.11,419(1413.3–1424.1) - 45OpenAIDossier
Arena rating
GPT-5.1, Arena rating: 1,416 (scale 1350.0 to 1500.1), 95% confidence interval 1407.2 to 1425.41,416(1407.2–1425.4) - 46Z.aiDossier
Arena rating
GLM-5V-Turbo, Arena rating: 1,406 (scale 1350.0 to 1500.1), 95% confidence interval 1398.0 to 1413.11,406(1398.0–1413.1)