All rankings / Search and documents

Search · style controlled

Human preference ratings for the overall category of this Arena task. Scores from different boards are not interchangeable.

42 entries · Arena score · Higher is better

Source publication: 2026-10-13. Retrieved 2026-11-05.

Source: Arena (lmarena-ai) · CC-BY-4.0.

Ranks use confidence bounds. Several models can share a rank; the rows within a shared rank are displayed by score.

Search · style controlled · Arena score
Significance rankModel / evaluated variantArena score95% confidence intervalVotes
1gpt-5.6-sol-xhighopenai1,257.931,251.46 – 1,264.460,437
1gpt-6.1-sol-maxopenai1,249.921,240.9 – 1,258.945,555
1gpt-6-astra-maxopenai1,246.241,238.2 – 1,254.2818,483
3claude-fable-5.1-highanthropic1,233.371,226.2 – 1,240.5412,782
2qwen3.8-maxalibaba1,229.51,216.2 – 1,242.791,870
4claude-fable-5-highanthropic1,229.021,222.44 – 1,235.673,596
4gpt-6-sol-maxopenai1,228.211,220.77 – 1,235.6511,313
4gpt-5.5-searchopenai1,226.451,221.46 – 1,231.45125,957
4claude-opus-4-6-searchanthropic1,222.051,217.36 – 1,226.73173,191
7claude-opus-5-highanthropic1,214.351,207.46 – 1,221.2413,285
8claude-opus-5.5-mediumanthropic1,211.231,204.13 – 1,218.3210,842
10claude-opus-4-7anthropic1,209.711,204.77 – 1,214.64127,506
9gpt-6-luna-maxopenai1,209.251,201.86 – 1,216.6510,793
10gemini-3.1-pro-groundinggoogle1,206.621,201.83 – 1,211.4151,436
10grok-4.20-multi-agent-beta-0309xai1,206.381,200.83 – 1,211.94126,238
10grok-4.5xai1,206.191,200.29 – 1,212.0976,458
11claude-sonnet-4-6-searchanthropic1,201.931,197.26 – 1,206.6173,986
11gemini-3-pro-groundinggoogle1,200.751,195.26 – 1,206.2438,857
11grok-4.20-beta1xai1,199.311,193.25 – 1,205.3857,049
13gpt-5.4-searchopenai1,196.261,190.59 – 1,201.94126,730
13o3-searchopenai1,196.161,189.43 – 1,202.920,770
15claude-opus-4-8anthropic1,195.251,189.61 – 1,200.89106,304
17grok-4-1-fast-searchxai1,1951,189.92 – 1,200.0885,919
17claude-sonnet-5-searchanthropic1,193.161,186.87 – 1,199.4572,978
17grok-4.3xai1,192.791,187.93 – 1,197.65128,393
15ernie-5.1baidu1,192.141,182.66 – 1,201.634,024
20gemini-3-flash-groundinggoogle1,188.681,184.28 – 1,193.07196,399
19gpt-5-searchopenai1,187.731,180.35 – 1,195.1120,865
20claude-opus-4-5-searchanthropic1,186.691,180.96 – 1,192.4265,047
24gpt-5.1-searchopenai1,183.641,178.27 – 1,189.0263,042
28claude-sonnet-4-5-searchanthropic1,175.981,170.74 – 1,181.21144,153
31claude-opus-4-1-searchanthropic1,167.091,162.11 – 1,172.0779,806
32gpt-5.2-searchopenai1,161.41,155.6 – 1,167.255,634
32grok-4-fast-searchxai1,161.081,156.26 – 1,165.8943,083
33claude-opus-4-searchanthropic1,151.841,146.29 – 1,157.3931,506
35gemini-2.5-pro-groundinggoogle1,144.231,139.43 – 1,149.0386,689
35ppl-sonar-pro-highperplexity1,143.221,137.32 – 1,149.1128,899
36ppl-sonar-reasoning-pro-highperplexity1,138.51,132.57 – 1,144.4329,439
36gpt-5.2-search-non-reasoningopenai1,135.761,129.95 – 1,141.5679,796
40grok-4-searchxai1,122.741,116.55 – 1,128.9219,195
41api-gpt-4o-searchopenai1,083.311,071.88 – 1,094.733,399
42diffbot-small-xldiffbot1,062.551,053.84 – 1,071.276,368

What this ranking can tell you

  • The population is the models in this source board, including variants and models without an available API price.
  • Ranks use the published confidence bounds: one plus the number of models whose lower bound exceeds this model’s upper bound. Shared ranks are not proof of equal performance.

Check the source

The original evidence is retained byte for byte. The downloadable board includes the source URLs, revision and SHA-256 hashes.

Download this board and provenance →

Source files and licences

Explore other model types →Read the benchmark guide →

Evidence & Ask