Unspecified benchmark
LM Arena Document Leaderboard
LM Arena Document is a current leaderboard using Arena Rating.
Updated Sep 8, 2026
Models33
Model coverage33
MetricArena Rating
EvidenceA
LM Arena Document Ranking
Higher arena rating ranks better on this benchmark.
Sort by
Rank | Model | Arena Rating | Percentile | Participants | Votes | Evidence | Evaluated |
|---|
| Rank01 | ModelAN | Arena Rating1,520.11 rating (1,505.48 rating - 1,534.73 rating) | Percentile100.00% | Participants33 | Votes1,663 | EvidenceA | Evaluated |
| Rank02 | ModelAN | Arena Rating1,509.94 rating (1,503.54 rating - 1,516.35 rating) | Percentile96.88% | Participants33 | Votes37,271 | EvidenceA | Evaluated |
| Rank03 | ModelAN | Arena Rating1,504.25 rating (1,495.08 rating - 1,513.42 rating) | Percentile93.75% | Participants33 | Votes5,792 | EvidenceA | Evaluated |
| Rank04 | ModelAN | Arena Rating1,498.15 rating (1,491.07 rating - 1,505.22 rating) | Percentile90.63% | Participants33 | Votes18,990 | EvidenceA | Evaluated |
| Rank05 | ModelOP | Arena Rating1,484.70 rating (1,477.69 rating - 1,491.71 rating) | Percentile87.50% | Participants33 | Votes16,816 | EvidenceA | Evaluated |
| Rank06 | ModelAN | Arena Rating1,482.80 rating (1,476.50 rating - 1,489.09 rating) | Percentile84.38% | Participants33 | Votes54,427 | EvidenceA | Evaluated |
| Rank07 | ModelOP | Arena Rating1,479.21 rating (1,466.19 rating - 1,492.22 rating) | Percentile81.25% | Participants33 | Votes1,969 | EvidenceA | Evaluated |
| Rank08 | ModelOP | Arena Rating1,478.68 rating (1,461.94 rating - 1,495.42 rating) | Percentile78.13% | Participants33 | Votes1,152 | EvidenceA | Evaluated |
| Rank09 | ModelAN | Arena Rating1,474.59 rating (1,466.45 rating - 1,482.73 rating) | Percentile75.00% | Participants33 | Votes8,388 | EvidenceA | Evaluated |
| Rank10 | ModelME | Arena Rating1,471.66 rating (1,458.81 rating - 1,484.51 rating) | Percentile71.88% | Participants33 | Votes2,054 | EvidenceA | Evaluated |
| Rank11 | ModelAN | Arena Rating1,470.22 rating (1,459.85 rating - 1,480.60 rating) | Percentile68.75% | Participants33 | Votes3,601 | EvidenceA | Evaluated |
| Rank12 | ModelOP | Arena Rating1,470.14 rating (1,463.54 rating - 1,476.73 rating) | Percentile65.63% | Participants33 | Votes29,809 | EvidenceA | Evaluated |
| Rank13 | ModelGO | Arena Rating1,464.65 rating (1,454.41 rating - 1,474.88 rating) | Percentile62.50% | Participants33 | Votes3,689 | EvidenceA | Evaluated |
| Rank14 | ModelOP | Arena Rating1,462.30 rating (1,449.02 rating - 1,475.59 rating) | Percentile59.38% | Participants33 | Votes1,888 | EvidenceA | Evaluated |
| Rank15 | ModelAN | Arena Rating1,461.65 rating (1,451.22 rating - 1,472.08 rating) | Percentile56.25% | Participants33 | Votes7,965 | EvidenceA | Evaluated |
| Rank16 | ModelXA | Arena Rating1,453.73 rating (1,441.94 rating - 1,465.53 rating) | Percentile53.13% | Participants33 | Votes2,435 | EvidenceA | Evaluated |
| Rank17 | ModelMA | Arena Rating1,450.90 rating (1,443.22 rating - 1,458.58 rating) | Percentile50.00% | Participants33 | Votes11,181 | EvidenceA | Evaluated |
| Rank18 | ModelAN | Arena Rating1,446.19 rating (1,439.60 rating - 1,452.79 rating) | Percentile46.88% | Participants33 | Votes28,634 | EvidenceA | Evaluated |
| Rank19 | ModelGO | Arena Rating1,445.32 rating (1,439.36 rating - 1,451.29 rating) | Percentile43.75% | Participants33 | Votes45,162 | EvidenceA | Evaluated |
| Rank20 | ModelME | Arena Rating1,443.25 rating (1,424.94 rating - 1,461.55 rating) | Percentile40.63% | Participants33 | Votes1,078 | EvidenceA | Evaluated |
| Rank21 | ModelAC | Arena Rating1,439.57 rating (1,428.96 rating - 1,450.18 rating) | Percentile37.50% | Participants33 | Votes3,088 | EvidenceA | Evaluated |
| Rank22 | ModelMI | Arena Rating1,434.82 rating (1,426.45 rating - 1,443.18 rating) | Percentile34.38% | Participants33 | Votes6,263 | EvidenceA | Evaluated |
| Rank23 | ModelGO | Arena Rating1,433.01 rating (1,424.11 rating - 1,441.92 rating) | Percentile31.25% | Participants33 | Votes10,739 | EvidenceA | Evaluated |
| Rank24 | ModelMA | Arena Rating1,429.49 rating (1,422.64 rating - 1,436.34 rating) | Percentile28.13% | Participants33 | Votes19,891 | EvidenceA | Evaluated |
| Rank25 | ModelGO | Arena Rating1,423.70 rating (1,415.51 rating - 1,431.88 rating) | Percentile25.00% | Participants33 | Votes10,577 | EvidenceA | Evaluated |
| Rank26 | ModelAN | Arena Rating1,423.03 rating (1,416.60 rating - 1,429.46 rating) | Percentile21.88% | Participants33 | Votes30,937 | EvidenceA | Evaluated |
| Rank27 | ModelGO | Arena Rating1,420.87 rating (1,414.58 rating - 1,427.16 rating) | Percentile18.75% | Participants33 | Votes24,963 | EvidenceA | Evaluated |
| Rank28 | ModelZA | Arena Rating1,417.18 rating (1,407.64 rating - 1,426.72 rating) | Percentile15.63% | Participants33 | Votes4,894 | EvidenceA | Evaluated |
| Rank29 | ModelXA | Arena Rating1,413.91 rating (1,406.62 rating - 1,421.20 rating) | Percentile12.50% | Participants33 | Votes18,666 | EvidenceA | Evaluated |
| Rank30 | ModelGO | Arena Rating1,413.32 rating (1,403.94 rating - 1,422.69 rating) | Percentile9.38% | Participants33 | Votes7,173 | EvidenceA | Evaluated |
LM Arena Document Highlights
The leading models and scores on this benchmark.
LM Arena Document Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for LM Arena Document
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is LM Arena Document?
What LM Arena Document measures and how its scores work.
LM Arena Document is a current leaderboard.
It reports Arena Rating in rating units.
Scores are shown in rating. This benchmark is verified and has an evidence level of A.
- Family
- lm-arena-document
- Modality
- Unspecified
- Primary category
- Unspecified
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- overall
LM Arena. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about LM Arena Document.
Which model scores highest on LM Arena Document?
Claude Opus 5 is currently ranked first with 1,520.11 rating.
What are the top three models on LM Arena Document?
The current leaders are Claude Opus 5 (1,520.11 rating), Claude Opus 4.6 (1,509.94 rating), and Claude Fable 5 (1,504.25 rating).
Which LM Arena Document model has the lowest official input price?
GPT-5.6 Luna has the lowest matched official input price at $0.20 input / $1.2 output per 1M tokens.
Which models are fastest among LM Arena Document results?
The fastest matched records are Gemini 3.5 Flash (210.94 tok/s via Google), GPT-5.5 Instant (206.89 tok/s via OpenAI), and GPT-5.5 (134.94 tok/s via OpenAI).
Does the highest arena rating result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.