Arena ranking
Browse Document results, model ratings and supporting selection context.
Data as of 2026-09-08
This source leaderboard ranking orders models by their original Arena rank rather than an aggregated benchmark score.
Rank | Model | Source Arena rating | LLMBoard Score | Votes | Official input / 1M | Output speed |
|---|
| Rank01 | ModelAN | Source Arena rating1,520.11 | LLMBoard ScoreN/A | Votes1,663 | Official input / 1M$5 | Output speed62.88 tok/s |
| Rank02 | ModelAN | Source Arena rating1,509.94 | LLMBoard ScoreN/A | Votes37,271 | Official input / 1M$5 | Output speed123.15 tok/s |
| Rank03 | ModelAN | Source Arena rating1,504.25 | LLMBoard ScoreN/A | Votes5,792 | Official input / 1M$10 | Output speed43.18 tok/s |
| Rank04 | ModelAN | Source Arena rating1,498.15 | LLMBoard ScoreN/A | Votes18,990 | Official input / 1M$5 | Output speed42.00 tok/s |
| Rank05 | ModelOP | Source Arena rating1,484.70 | LLMBoard ScoreN/A | Votes16,816 | Official input / 1M$5 | Output speed134.94 tok/s |
| Rank06 | ModelAN | Source Arena rating1,482.80 | LLMBoard ScoreN/A | Votes54,427 | Official input / 1M$3 | Output speed7.67 tok/s |
| Rank07 | ModelOP | Source Arena rating1,479.21 | LLMBoard ScoreN/A | Votes1,969 | Official input / 1M$2 | Output speed99.35 tok/s |
| Rank08 | ModelOP | Source Arena rating1,478.68 | LLMBoard ScoreN/A | Votes1,152 | Official input / 1M$4 | Output speed2.36 tok/s |
| Rank09 | ModelAN | Source Arena rating1,474.59 | LLMBoard ScoreN/A | Votes8,388 | Official input / 1M$5 | Output speed42.00 tok/s |
| Rank10 | ModelME | Source Arena rating1,471.66 | LLMBoard ScoreN/A | Votes2,054 | Official input / 1M$1.25 | Output speed14.00 tok/s |
| Rank11 | ModelAN | Source Arena rating1,470.22 | LLMBoard ScoreN/A | Votes3,601 | Official input / 1M$2 | Output speed42.00 tok/s |
| Rank12 | ModelOP | Source Arena rating1,470.14 | LLMBoard ScoreN/A | Votes29,809 | Official input / 1M$2.5 | Output speed50.00 tok/s |
| Rank13 | ModelGO | Source Arena rating1,464.65 | LLMBoard ScoreN/A | Votes3,689 | Official input / 1M$1.5 | Output speed210.94 tok/s |
| Rank14 | ModelOP | Source Arena rating1,462.30 | LLMBoard ScoreN/A | Votes1,888 | Official input / 1M$0.20 | Output speed41.87 tok/s |
| Rank15 | ModelAN | Source Arena rating1,461.65 | LLMBoard ScoreN/A | Votes7,965 | Official input / 1M$5 | Output speed42.00 tok/s |
| Rank16 | ModelXA | Source Arena rating1,453.73 | LLMBoard ScoreN/A | Votes2,435 | Official input / 1M$2 | Output speed80.00 tok/s |
| Rank17 | ModelMA | Source Arena rating1,450.90 | LLMBoard ScoreN/A | Votes11,181 | Official input / 1M$0.95 | Output speed1.24 tok/s |
| Rank18 | ModelAN | Source Arena rating1,446.19 | LLMBoard ScoreN/A | Votes28,634 | Official input / 1M$3 | Output speed42.00 tok/s |
| Rank19 | ModelGO | Source Arena rating1,445.32 | LLMBoard ScoreN/A | Votes45,162 | Official input / 1M$2 | Output speed43.79 tok/s |
| Rank20 | ModelME | Source Arena rating1,443.25 | LLMBoard ScoreN/A | Votes1,078 | Official input / 1MN/A | Output speedN/A |
| Rank21 | ModelAC | Source Arena rating1,439.57 | LLMBoard ScoreN/A | Votes3,088 | Official input / 1M$0.50 | Output speedN/A |
| Rank22 | ModelMI | Source Arena rating1,434.82 | LLMBoard ScoreN/A | Votes6,263 | Official input / 1M$0.30 | Output speed6.61 tok/s |
| Rank23 | ModelGO | Source Arena rating1,433.01 | LLMBoard ScoreN/A | Votes10,739 | Official input / 1MN/A | Output speed90.00 tok/s |
| Rank24 | ModelMA | Source Arena rating1,429.49 | LLMBoard ScoreN/A | Votes19,891 | Official input / 1MN/A | Output speedN/A |
| Rank25 | ModelGO | Source Arena rating1,423.70 | LLMBoard ScoreN/A | Votes10,577 | Official input / 1MN/A | Output speed22.86 tok/s |
| Rank26 | ModelAN | Source Arena rating1,423.03 | LLMBoard ScoreN/A | Votes30,937 | Official input / 1M$1 | Output speed100.00 tok/s |
| Rank27 | ModelGO | Source Arena rating1,420.87 | LLMBoard ScoreN/A | Votes24,963 | Official input / 1M$1.25 | Output speed85.00 tok/s |
| Rank28 | ModelZA | Source Arena rating1,417.18 | LLMBoard ScoreN/A | Votes4,894 | Official input / 1M$5 | Output speedN/A |
| Rank29 | ModelXA | Source Arena rating1,413.91 | LLMBoard ScoreN/A | Votes18,666 | Official input / 1MN/A | Output speedN/A |
| Rank30 | ModelGO | Source Arena rating1,413.32 | LLMBoard ScoreN/A | Votes7,173 | Official input / 1M$0.50 | Output speed124.32 tok/s |
This leaderboard overview compares the source ranking, rating gaps and voting activity.
Use intervals and vote counts to judge how firmly each leaderboard position is supported; source ratings remain separate from the benchmark ranking.
Vendor representation among the top-ranked models in this leaderboard.
A practical summary of the first five entries in this document AI model leaderboard, with benchmark evidence and pricing kept in context.
Common questions about the Document Leaderboard.
Claude Opus 5 is currently ranked first with 1,520.11 Rating.
The current leaders are Claude Opus 5 (rank #1), Claude Opus 4.6 (rank #2), and Claude Fable 5 (rank #3).
GPT-5.6-Luna has the lowest matched official input price at $0.2 per 1M tokens.
The fastest matched records are GPT-5.1 (369.03 tok/s via OpenAI), Gemini 3.5 Flash (210.94 tok/s via Google), and GPT-5.5-Instant (206.89 tok/s via OpenAI).
No. Source Arena results describe a particular task or preference signal. Capability benchmarks, prices and runtime can produce different rankings.
This source-native ranking follows the original source metric. Benchmark evidence is shown separately and is not substituted for the source result.
This page currently compares 33 models.
A model may not yet have a matching Arena result, or multiple version results may resolve to the same model.
Ranking basisThis document AI model leaderboard uses the source Arena rating and rank. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
Selection summary
Claude Opus 5 currently leads the document ranking at 1,520.11. Compare evidence coverage, source activity and price separately before choosing a model for production.
Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.