Benchmark

Document Arena (arena.ai)

Measures: Users upload documents and ask two anonymous models to analyse or reason over the long content, then vote; votes are turned into an Elo-scale score with a confidence interval (Style Control off by default).

Maintainer
Arena Intelligence (arena.ai; formerly LMArena / LMSYS Chatbot Arena)
Introduced
—
Status
Active
Saturation
Headroom
Typical provenance
Independent
In scoring set
Yes

Official leaderboard

Leaderboard

Independent and third-party results only.

Leaderboard
ModelArena score (Elo scale)Score CI (+/-)VotesProvenanceSourceEvaluated
GLM-5V-Turbo1400—6995IndependentArena13 Sep 2026
GPT-5.11416—8244IndependentArena13 Sep 2026
Claude Haiku 4.51419—34677IndependentArena13 Sep 2026
GPT-5.2
OpenAI · High effort
1423—7108IndependentArena13 Sep 2026
GPT-5.21425—28212IndependentArena13 Sep 2026
Gemini 2.5 Pro1431—25110IndependentArena13 Sep 2026
MiniMax-M31433—6314IndependentArena13 Sep 2026
Kimi K2.5
Moonshot AI · Thinking effort
1436—21569IndependentArena13 Sep 2026
Gemini 3 Flash1439—7158IndependentArena13 Sep 2026
Grok 4.20
SpaceXAI (xAI) · Reasoning effort
1439—20948IndependentArena13 Sep 2026
Gemma 4 31B1445—12326IndependentArena13 Sep 2026
Claude Sonnet 4.51449—31455IndependentArena13 Sep 2026
Kimi K2.61450—11291IndependentArena13 Sep 2026
Gemini 3 Pro1451—10769IndependentArena13 Sep 2026
Qwen3.7-Plus1453—4591IndependentArena13 Sep 2026
Gemini 3.5 Flash
Google DeepMind · High effort
1453—4061IndependentArena13 Sep 2026
Gemini 3.6 Flash
Google DeepMind · High effort
1453—2975IndependentArena13 Sep 2026
Gemini 3.1 Pro1459—49457IndependentArena13 Sep 2026
Gemini 3.5 Flash
Google DeepMind · Medium effort
1461—6500IndependentArena13 Sep 2026
Grok 4.6
SpaceXAI (xAI) · High effort
1461—2258IndependentArena13 Sep 2026
GPT-5.6 Luna
OpenAI · Xhigh effort
1462—5766IndependentArena13 Sep 2026
Grok 4.51464—4965IndependentArena13 Sep 2026
Claude Opus 4.51465—7981IndependentArena13 Sep 2026
Muse Spark 1.3
Meta · Max effort
1468—1006IndependentArena13 Sep 2026
GPT-5.41470—33331IndependentArena13 Sep 2026
GPT-6 Astra
OpenAI · Max effort
1473—1621IndependentArena13 Sep 2026
Claude Opus 4.81475—12128IndependentArena13 Sep 2026
Claude Sonnet 5
Anthropic · High effort
1476—7411IndependentArena13 Sep 2026
GPT-5.6 Terra
OpenAI · Xhigh effort
1477—5787IndependentArena13 Sep 2026
Claude Sonnet 4.61478—57813IndependentArena13 Sep 2026
Muse Spark 1.11479—4885IndependentArena13 Sep 2026
GPT-5.5
OpenAI · High effort
1480—20944IndependentArena13 Sep 2026
GPT-5.51483—21279IndependentArena13 Sep 2026
Claude Opus 4.8
Anthropic · High effort
1486—11551IndependentArena13 Sep 2026
GPT-5.6 Sol
OpenAI · Xhigh effort
1489—4818IndependentArena13 Sep 2026
Claude Opus 5
Anthropic · High effort
1490—8794IndependentArena13 Sep 2026
Claude Fable 5.1
Anthropic · Max effort
1493—1403IndependentArena13 Sep 2026
Claude Opus 4.7
Anthropic · High effort
1493—21957IndependentArena13 Sep 2026
Claude Opus 4.61494—41260IndependentArena13 Sep 2026
Claude Opus 4.71494—22192IndependentArena13 Sep 2026
Claude Opus 4.6
Anthropic · High effort
1495—27929IndependentArena13 Sep 2026
Claude Fable 51499—8497IndependentArena13 Sep 2026
  • Arena (arena.ai), lmarena-ai/leaderboard-dataset, CC BY 4.0.

Sources