Boards
Ranked by a named score.
Voice has no licensed leaderboard in this snapshot, so it is not given a made-up ranking. Open the directory and filter to voice.
Reasoning
LMArena text overall rating. Higher is the board's own Elo-style rating. LMArena dataset is CC-BY-4.0. Epoch is CC-BY-4.0.
| Rank | Model | Score | Votes |
|---|---|---|---|
| 1 | Gemini 4 Argon (High) google | 1532.3 | 4,900 |
| 2 | Claude Opus 5.5 (High) anthropic | 1516.2 | 9,496 |
| 3 | Claude Fable 5.1 (Max) anthropic | 1511.9 | 15,623 |
| 4 | Claude Opus 5 (Max) anthropic | 1508.6 | 33,683 |
| 5 | Claude Opus 5 (High) anthropic | 1502.8 | 69,307 |
| 6 | Muse Wonder (Max) meta | 1501.7 | 5,099 |
| 7 | Claude Opus 4.6 (High) anthropic | 1501.2 | 82,619 |
| 8 | Gemini 3.8 Flash (High) google | 1498.9 | 41,423 |
| 9 | Claude Opus 4.6 anthropic | 1497.0 | 87,372 |
| 10 | Claude Fable 5 (High) anthropic | 1491.3 | 43,399 |
| 11 | MiMo-V2.6-Pro xiaomi | 1491.1 | 4,065 |
| 12 | Muse Spark 1.3 (Max) meta | 1490.5 | 19,323 |
| 13 | Claude Opus 4.7 (High) anthropic | 1489.7 | 65,131 |
| 14 | Gemini 3.7 Flash (High) google | 1488.5 | 29,177 |
| 15 | Qwen3.8-Max alibaba | 1483.2 | 29,418 |
| 16 | Claude Opus 4.7 anthropic | 1483.0 | 66,185 |
| 17 | Muse Spark 1.2 (xHigh) meta | 1482.4 | 9,578 |
| 18 | Gemini 3.5 Flash (High) google | 1482.0 | 48,479 |
| 19 | Gemini 3.1 Pro Preview google | 1480.1 | 129,523 |
| 20 | Gemini 3.6 Flash (High) google | 1479.5 | 36,079 |
| 21 | Gemini 3 Pro google | 1479.5 | 42,059 |
| 22 | Muse Spark 1.1 meta | 1478.3 | 44,815 |
| 23 | Gemini 3.5 Flash (Medium) google | 1477.6 | 47,228 |
| 24 | Kimi K3 (Max) moonshot | 1475.7 | 34,325 |
| 25 | Qwen3.7-Max-Preview alibaba | 1474.9 | 3,942 |
Coding
LMArena text category 'coding'. LMArena dataset is CC-BY-4.0. Epoch is CC-BY-4.0.
| Rank | Model | Score | Votes |
|---|---|---|---|
| 1 | Gemini 4 Argon (High) google | 1546.0 | 1,236 |
| 2 | Claude Opus 5.5 (High) anthropic | 1545.0 | 2,143 |
| 3 | Claude Opus 4.6 anthropic | 1534.5 | 24,104 |
| 4 | Claude Opus 5 (Max) anthropic | 1533.2 | 8,413 |
| 5 | MiMo-V2.6-Pro xiaomi | 1533.2 | 1,102 |
| 6 | Claude Fable 5.1 (Max) anthropic | 1532.5 | 3,112 |
| 7 | Claude Opus 5 (High) anthropic | 1532.3 | 17,823 |
| 8 | Claude Opus 4.6 (High) anthropic | 1530.7 | 21,348 |
| 9 | Muse Wonder (Max) meta | 1522.1 | 1,234 |
| 10 | Claude Opus 4.7 (High) anthropic | 1518.5 | 18,580 |
| 11 | Claude Fable 5 (High) anthropic | 1518.3 | 11,128 |
| 12 | Claude Opus 4.7 anthropic | 1514.8 | 18,786 |
| 13 | Claude Sonnet 5.5 (xHigh) anthropic | 1513.1 | 1,914 |
| 14 | Muse Spark 1.3 (Max) meta | 1510.2 | 4,958 |
| 15 | Gemini 3.8 Flash (High) google | 1508.3 | 10,731 |
| 16 | Kimi K3 (Max) moonshot | 1508.3 | 8,643 |
| 17 | GLM-5.3-Flash zai | 1506.8 | 9,192 |
| 18 | Muse Spark 1.2 (xHigh) meta | 1505.2 | 2,474 |
| 19 | Claude Opus 4.5 (High, 32K, 20251101) anthropic | 1503.5 | 7,787 |
| 20 | Claude Sonnet 4.6 anthropic | 1503.5 | 19,732 |
| 21 | DeepSeek-V4.1-Flash-20260910 (Max) deepseek | 1502.1 | 3,680 |
| 22 | Qwen3.8-Max alibaba | 1501.6 | 8,078 |
| 23 | MiMo-V2.5-Pro xiaomi | 1501.1 | 20,773 |
| 24 | MiMo-V2.6-Flash xiaomi | 1499.1 | 3,010 |
| 25 | Gemini 3.7 Flash (High) google | 1498.9 | 7,905 |
Math
LMArena text category 'math'. LMArena dataset is CC-BY-4.0. Epoch is CC-BY-4.0.
| Rank | Model | Score | Votes |
|---|---|---|---|
| 1 | Gemini 4 Argon (High) google | 1536.4 | 273 |
| 2 | Claude Opus 5 (Max) anthropic | 1533.1 | 1,597 |
| 3 | Gemini 3.8 Flash (High) google | 1532.0 | 2,094 |
| 4 | Claude Opus 5 (High) anthropic | 1530.9 | 3,315 |
| 5 | Claude Fable 5.1 (Max) anthropic | 1530.8 | 746 |
| 6 | Claude Opus 5.5 (High) anthropic | 1517.7 | 491 |
| 7 | Claude Opus 4.6 (High) anthropic | 1517.6 | 4,248 |
| 8 | Claude Fable 5 (High) anthropic | 1516.5 | 2,161 |
| 9 | Gemini 3.7 Flash (High) google | 1510.9 | 1,501 |
| 10 | Claude Opus 4.6 anthropic | 1509.5 | 4,741 |
| 11 | Gemini 3.6 Flash (High) google | 1507.5 | 1,760 |
| 12 | Gemini 3.5 Flash (High) google | 1504.2 | 2,388 |
| 13 | GLM-5.3-Flash zai | 1502.7 | 1,820 |
| 14 | Qwen3.8-Max alibaba | 1501.4 | 1,481 |
| 15 | MiMo-V2.6-Pro xiaomi | 1499.9 | 215 |
| 16 | Claude Opus 4.7 (High) anthropic | 1498.5 | 3,293 |
| 17 | Muse Spark 1.3 (Max) meta | 1496.9 | 948 |
| 18 | Claude Sonnet 5.5 (xHigh) anthropic | 1496.7 | 416 |
| 19 | GLM-5.3 (Max) zai | 1495.4 | 1,023 |
| 20 | Kimi K3 (Max) moonshot | 1493.8 | 1,535 |
| 21 | Claude Opus 4.7 anthropic | 1489.3 | 3,389 |
| 22 | GPT-6.1 Sol (Max) openai | 1488.9 | 390 |
| 23 | Qwen3.7-Max-Preview alibaba | 1487.9 | 233 |
| 24 | GPT-5.4 (High) openai | 1487.5 | 3,377 |
| 25 | GPT-5.5 openai | 1487.0 | 3,518 |
Image
LMArena text-to-image board. LMArena dataset is CC-BY-4.0. Epoch is CC-BY-4.0.
| Rank | Model | Score | Votes |
|---|---|---|---|
| 1 | gpt-image-2.5-sunburst openai | 1424.5 | 17,752 |
| 2 | gpt-image-2.5-flare openai | 1397.6 | 16,889 |
| 3 | gpt-image-2 (medium) openai | 1383.0 | 93,334 |
| 4 | grok-imagine-image-2.0 (canvas) xai | 1333.5 | 3,113 |
| 5 | mai-image-2.6 microsoft-ai | 1332.1 | 25,091 |
| 6 | gemini-nano-banana-2.1 google | 1327.8 | 5,310 |
| 7 | reve-2.1 reve | 1301.7 | 8,249 |
| 8 | grok-imagine-image-2.0 (20260801) xai | 1294.2 | 14,483 |
| 9 | muse-image meta | 1274.0 | 42,262 |
| 10 | reve-2.0 reve | 1269.2 | 15,773 |
| 11 | gemini-3.1-flash-image (nano-banana-2) [web-search] google | 1261.8 | 60,515 |
| 12 | seedream-5.0-pro bytedance | 1255.5 | 94,181 |
| 13 | qwen-image-3.0-pro alibaba | 1255.2 | 11,440 |
| 14 | mai-image-2.5 microsoft-ai | 1254.2 | 61,265 |
| 15 | gemini-3.1-flash-lite-image (nano-banana-2-lite) google | 1250.2 | 17,804 |
| 16 | gemini-3-pro-image-2k (nano-banana-pro) google | 1247.9 | 175,734 |
| 17 | gpt-image-1.5-high-fidelity openai | 1237.4 | 169,927 |
| 18 | gemini-3-pro-image-preview (nano-banana-pro) google | 1234.1 | 86,464 |
| 19 | qwen-image-2.1 alibaba | 1222.9 | 11,441 |
| 20 | ideogram-4.0-quality ideogram | 1205.1 | 60,591 |
| 21 | qwen-image-2.0-pro-2026-06-22 alibaba | 1191.8 | 13,096 |
| 22 | uni-1.1-max luma-ai | 1189.3 | 14,072 |
| 23 | uni-1.1 luma-ai | 1184.3 | 46,699 |
| 24 | mai-image-2 microsoft-ai | 1183.8 | 51,475 |
| 25 | grok-imagine-image xai | 1169.7 | 272,085 |
Video
LMArena image-to-video board. LMArena dataset is CC-BY-4.0. Epoch is CC-BY-4.0.
| Rank | Model | Score | Votes |
|---|---|---|---|
| 1 | minimax-h3 minimax | 1495.4 | 57,112 |
| 2 | gemini-omni-1.1-flash google | 1488.5 | 3,734 |
| 3 | wan3.0 alibaba | 1479.8 | 3,442 |
| 4 | dreamina-seedance-2.5-720p bytedance | 1477.4 | 15,729 |
| 5 | dreamina-seedance-2.0-720p bytedance | 1474.6 | 121,427 |
| 6 | gemini-omni-flash google | 1464.8 | 87,155 |
| 7 | hidream-o1-video-1.0 hidream | 1456.5 | 6,488 |
| 8 | grok-imagine-video-1.5-720p xai | 1456.0 | 137,459 |
| 9 | flux-3-video-20260811 bfl | 1449.5 | 37,042 |
| 10 | happyhorse-1.0 aorizon | 1441.7 | 70,640 |
| 11 | wan2.7-i2v wan | 1428.2 | 85,521 |
| 12 | grok-imagine-video-720p xai | 1415.3 | 554,050 |
| 13 | veo-3.1-audio google | 1398.1 | 25,116 |
| 14 | veo-3.1-audio-1080p google | 1390.2 | 53,089 |
| 15 | veo-3.1-fast-audio google | 1385.1 | 99,804 |
| 16 | grok-imagine-video-480p xai | 1384.9 | 19,429 |
| 17 | veo-3.1-fast-audio-1080p google | 1371.6 | 54,380 |
| 18 | vidu-q3-pro shengshu | 1363.2 | 36,679 |
| 19 | kling-v3-pro kling | 1354.0 | 206,221 |
| 20 | veo-3-audio google | 1331.5 | 32,367 |
| 21 | veo-3-fast-audio google | 1326.0 | 41,194 |
| 22 | wan2.5-i2v-preview alibaba | 1321.4 | 16,907 |
| 23 | wan2.6-i2v alibaba | 1309.5 | 121,082 |
| 24 | seedance-v1.5-pro bytedance | 1307.2 | 245,206 |
| 25 | pixverse-v5.6 | 1299.3 | 126,710 |
GPQA Diamond
Epoch AI, percent. CC-BY-4.0. Not mixed into the Value Index. LMArena dataset is CC-BY-4.0. Epoch is CC-BY-4.0.
| Rank | Model | Score | Votes |
|---|---|---|---|
| 1 | Gemini 4 Argon Google | 96.4% | — |
| 2 | GPT-6 Astra OpenAI | 95.8% | — |
| 3 | Claude Sonnet 5.5 Anthropic | 95.6% | — |
| 4 | GPT-6.1 Sol OpenAI | 95.4% | — |
| 5 | Gemini 3.8 Flash Google | 95.4% | — |
| 6 | Gemini 3.7 Flash Google | 94.8% | — |
| 7 | GPT-6 Sol OpenAI | 94.3% | — |
| 8 | Gemini 3.6 Flash Google | 94.1% | — |
| 9 | Gemini 3.1 Pro Preview Google | 94.1% | — |
| 10 | Grok 4.6 xAI | 94.0% | — |
| 11 | Claude Opus 5 Anthropic | 93.9% | — |
| 12 | GPT-5.6 Sol OpenAI | 93.5% | — |
| 13 | Grok 4.5 xAI | 93.4% | — |
| 14 | GPT-5.6 Terra OpenAI | 93.3% | — |
| 15 | DeepSeek V4.1 Flash 1017 DeepSeek | 93.2% | — |
| 16 | Kimi K3 Nvidia | 93.1% | — |
| 17 | Gemini 3.5 Flash Google | 92.8% | — |
| 18 | Grok 4.7 xAI | 92.7% | — |
| 19 | Qwen3.8 Max Alibaba | 92.7% | — |
| 20 | GLM-5.2 Alibaba | 91.9% | — |
| 21 | GPT-5.6 Luna OpenAI | 91.6% | — |
| 22 | Claude Opus 4.8 Anthropic | 91.0% | — |
| 23 | DeepSeek V4 Flash 0731 Alibaba | 91.0% | — |
| 24 | GLM-5.3 Nvidia | 90.9% | — |
| 25 | Qwen3.7 Max Alibaba | 90.9% | — |