Arena benchmark
LM Arena Agent Leaderboard
LM Arena Agent model ratings and results.
Updated Aug 17, 2026. Data as of 2026-08-17
Models40
Categories1
Result typeagent_score
Agent Ranking
Models are ordered by their Arena rank.
Sort by
Rank | Model | Organization | Rating / score | Votes | Observations | Result date |
|---|
| Rank01 | ModelAN | Organizationanthropic | Rating / score0.1 | VotesN/A | Observations2.2M | Result date |
| Rank02 | ModelAN | Organizationanthropic | Rating / score0.1 | VotesN/A | Observations911.7K | Result date |
| Rank04 | ModelOP | Organizationopenai | Rating / score0.1 | VotesN/A | Observations1.6M | Result date |
| Rank05 | ModelMA | Organizationmoonshot | Rating / score0.1 | VotesN/A | Observations2.4M | Result date |
| Rank07 | ModelOP | Organizationopenai | Rating / score0.1 | VotesN/A | Observations2M | Result date |
| Rank10 | ModelAN | Organizationanthropic | Rating / score0.1 | VotesN/A | Observations1.4M | Result date |
| Rank11 | ModelAC | Organizationalibaba | Rating / score0.1 | VotesN/A | Observations561.8K | Result date |
| Rank12 | ModelAN | Organizationanthropic | Rating / score0.1 | VotesN/A | Observations2.5M | Result date |
| Rank13 | ModelAN | Organizationanthropic | Rating / score0.1 | VotesN/A | Observations1.3M | Result date |
| Rank14 | ModelZA | Organizationzai | Rating / score0.1 | VotesN/A | Observations2.3M | Result date |
| Rank16 | ModelXA | Organizationxai | Rating / score0.1 | VotesN/A | Observations1.4M | Result date |
| Rank17 | ModelOP | Organizationopenai | Rating / score0.0 | VotesN/A | Observations3M | Result date |
| Rank18 | ModelOP | Organizationopenai | Rating / score0.0 | VotesN/A | Observations528.2K | Result date |
| Rank19 | ModelDE | Organizationdeepseek | Rating / score0.0 | VotesN/A | Observations3.2M | Result date |
| Rank20 | ModelOP | Organizationopenai | Rating / score0.0 | VotesN/A | Observations744.6K | Result date |
| Rank22 | ModelAN | Organizationanthropic | Rating / score0.0 | VotesN/A | Observations1.3M | Result date |
| Rank23 | ModelAN | Organizationanthropic | Rating / score0.0 | VotesN/A | Observations1.4M | Result date |
| Rank24 | ModelME | Organizationmeta | Rating / score0.0 | VotesN/A | Observations2.5M | Result date |
| Rank25 | ModelMA | Organizationmoonshot | Rating / score0.0 | VotesN/A | Observations402.4K | Result date |
| Rank26 | ModelZA | Organizationzai | Rating / score0.0 | VotesN/A | Observations3.2M | Result date |
| Rank27 | ModelAC | Organizationalibaba | Rating / score0.0 | VotesN/A | Observations1.3M | Result date |
| Rank28 | ModelDE | Organizationdeepseek | Rating / score0.0 | VotesN/A | Observations1.4M | Result date |
| Rank29 | ModelGO | Organizationgoogle | Rating / score-0.0 | VotesN/A | Observations4M | Result date |
| Rank30 | ModelGO | Organizationgoogle | Rating / score-0.0 | VotesN/A | Observations2.6M | Result date |
| Rank31 | ModelMA | Organizationmoonshot | Rating / score-0.0 | VotesN/A | Observations321.5K | Result date |
| Rank32 | ModelTE | Organizationtencent | Rating / score-0.0 | VotesN/A | Observations620.8K | Result date |
| Rank33 | ModelXI | Organizationxiaomi | Rating / score-0.0 | VotesN/A | Observations1.3M | Result date |
| Rank34 | ModelAC | Organizationalibaba | Rating / score-0.0 | VotesN/A | Observations699.4K | Result date |
| Rank35 | ModelDE | Organizationdeepseek | Rating / score-0.0 | VotesN/A | Observations1.2M | Result date |
| Rank37 | ModelMI | Organizationminimax | Rating / score-0.0 | VotesN/A | Observations2M | Result date |
| Rank39 | ModelTHInklingThinkingmachines | Organizationthinky | Rating / score-0.1 | VotesN/A | Observations1.3M | Result date |
| Rank40 | ModelMA | Organizationmistral | Rating / score-0.1 | VotesN/A | Observations168.1K | Result date |
| Rank41 | ModelXA | Organizationxai | Rating / score-0.1 | VotesN/A | Observations1.6M | Result date |
| Rank42 | ModelGO | Organizationgoogle | Rating / score-0.1 | VotesN/A | Observations1.9M | Result date |
| Rank43 | ModelXA | Organizationxai | Rating / score-0.1 | VotesN/A | Observations5.5M | Result date |
| Rank44 | ModelUP | Organizationupstage | Rating / score-0.1 | VotesN/A | Observations165.8K | Result date |
| Rank45 | ModelGO | Organizationgoogle | Rating / score-0.1 | VotesN/A | Observations551.6K | Result date |
| Rank46 | ModelMI | Organizationminimax | Rating / score-0.1 | VotesN/A | Observations668K | Result date |
| Rank47 | ModelNV | Organizationnvidia | Rating / score-0.1 | VotesN/A | Observations160.2K | Result date |
| Rank49 | ModelGO | Organizationgoogle | Rating / score-0.2 | VotesN/A | Observations708.1K | Result date |
Agent Rating Distribution
A closer view of the leading model ratings.
The Top AI Models for Agent
The first five models in this source ranking, with price and output speed included when a canonical model match is available.
Ranking basisThis agent AI model leaderboard uses the source agent score and Arena rank. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
- 01ANClaude Opus 5Anthropic
- agent score
- 0.12
- Price
- $5.0 input / $25 output per 1M tokens
- Speed
- Up to 3.3 tok/s via Anthropic
Strengths
- Ranks #1 on Agent
- 2,246,557 observations support the result
- Matched official PAYG token pricing is available
Considerations
- Arena results are independent from the LLMBoard capability score
- 02ANClaude Fable 5Anthropic
- agent score
- 0.12
- Price
- $10 input / $50 output per 1M tokens
- Speed
- Up to 9.1 tok/s via Anthropic
Strengths
- Ranks #2 on Agent
- 911,709 observations support the result
- Matched official PAYG token pricing is available
Considerations
- Arena results are independent from the LLMBoard capability score
- 04OPGPT-5.6 SolOpenAI
- agent score
- 0.11
- Price
- $5.0 input / $30 output per 1M tokens
- Speed
- Up to 27 tok/s via OpenAI
Strengths
- Ranks #4 on Agent
- 1,599,076 observations support the result
- Matched official PAYG token pricing is available
Considerations
- Arena results are independent from the LLMBoard capability score
- 05MAKimi K3Moonshot AI
- agent score
- 0.11
- Price
- $3.0 input / $15 output per 1M tokens
- Speed
- Up to 26 tok/s via Fireworks
Strengths
- Ranks #5 on Agent
- 2,431,932 observations support the result
- Matched official PAYG token pricing is available
Considerations
- Arena results are independent from the LLMBoard capability score
- 07OPGPT-5.5OpenAI
- agent score
- 0.09
- Price
- $5.0 input / $30 output per 1M tokens
Strengths
- Ranks #7 on Agent
- 2,028,364 observations support the result
- Matched official PAYG token pricing is available
Considerations
- Arena results are independent from the LLMBoard capability score
Selection summary
Best AI Models for Agent
Claude Opus 5 currently leads Agent at 0.12. The best AI model for this use case may change when price, speed and independent benchmark capability are considered.
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.
What is the Agent Arena?
What Agent measures and how to read its results.
Agent reports agent score values and currently defines 1 evaluation category.
Arena results are preference or task-outcome signals. They are not automatically equivalent to capability benchmark scores.
FAQ
Common questions about Agent.
Which model leads Agent?
Claude Opus 5 is currently ranked first.
What does Agent measure?
This Arena reports agent score values across 1 configured evaluation category.
Does this result affect the LLMBoard score?
No. Arena results remain a separate preference or task-outcome signal.
How many models are included?
40 model results are currently shown.