Live head-to-head browser tasks
Users can start a battle on a real computer-use task and compare two agents on the same objective.
Coarena
Freemium
Coarena é uma arena de uso do computador para publicar tarefas reais no navegador, acompanhar ao vivo dois agentes de IA de ponta tentando executá-las e julgar o vencedor às cegas. Também oferece métricas de benchmark, rótulos de preferência e acesso a datasets para avaliação.
O que é Coarena?
Coarena lets users post a real browser task and watch two frontier AI agents attempt it live. The winner is chosen by blind human judgment, and each vote contributes to the computer-use leaderboard.
The product also extends beyond the arena format. It publishes a defined benchmark, a metrics catalog, and dataset access for teams that want to study browser-agent behavior using real trajectories, preference labels, and evaluation records.
O que Coarena pode fazer?
Blind human voting
Judges pick the winner without being told which model produced which run, which keeps the comparison focused on observed performance.
Published computer-use benchmark
The benchmark page defines 57 measures across outcome, route, recovery, precision, tempo, cost, expression, output, head-to-head, and human judgment.
Metrics grounded in trajectories
Reported numbers are derived from real browser trajectories, including actions, coordinates, page context, and browser replies, rather than from a simple score alone.
Dataset and preference-label access
The data page describes licensed training data from live battles, including complete trajectories and blind pairwise preference labels.
Rule-linked publishing
The site states that numbers are published with the rule that produced them, helping readers interpret each metric in context.
Casos de uso
“Compare computer-use agents on the same task”
Run two agents against the same browser job and use blind judging to see which one finishes better in practice.
“Study evaluation metrics for agent research”
Use the benchmark definitions to understand how Coarena measures completion, recovery, precision, cost, and judgment quality.
“Source training or preference data”
Teams working on browser agents can request access to licensed trajectory, preference, or eval data described on the data page.
“Inspect how agents handle failure and recovery”
Researchers can use the metric catalog to examine not only whether a task was completed, but also how an agent responded when a step failed.
“Audit model behavior on real workflows”
Because the arena is based on actual browser work, it is suited to evaluating practical tasks rather than synthetic prompts alone.
Perguntas frequentes
How does Coarena choose a winner?
The arena uses blind human judgment. Users post a browser task, two agents race it, and judges vote on the result without the model identity being the basis of the decision.
What kind of metrics does the benchmark publish?
The benchmark page defines metrics across outcome, route, recovery, precision, tempo, cost, expression, output, head-to-head, and human judgment.
Does Coarena provide dataset access?
Yes. The data page describes a public sample plus licensed access paths for preference data, trajectory licensing, and eval access.
Are screenshots included in the dataset?
No. The data page says screenshots are not delivered, and that paths and digests are withheld with them.
Is pricing available on the site?
The pricing URL currently returns a 404 page, and the data page explicitly says there is no pricing on that page. Public pricing details were not evidenced in the supplied sources.
Informações rápidas
- Category
- Computer-use evaluation
- Product type
- Arena, benchmark, and dataset platform
- Primary workflow
- Post a browser task, watch two agents run it, and vote blind
- Benchmark size
- 57 defined measures
- Data access
- Public sample plus request-access tiers
- Domain
- coarena.ai
Análise de tráfego de Coarena
Os dados de tráfego são apenas para referência.
- Visitas mensais
- 0
- Classificação global
- -
- Taxa de rejeição
- 0.00%
- Duração média da visita
- 00:00
- Páginas por visita
- 0.00
- Classificação do domínio
- 11
Tendências de tráfego
| mai. | 0 |
|---|---|
| jun. | 0 |
| jul. | 0 |
Fontes de tráfego
As análises de tráfego ainda não estão disponíveis.
Principais regiões
As análises de tráfego ainda não estão disponíveis.
Alternativas ao Coarena
Kitesurf
developers.cloudflare.com
Kitesurf é o navegador sem estado da Cloudflare para agentes de IA e fluxos de trabalho automatizados. Projetado para funcionar no Workers, renderiza páginas, captura telas e extrai HTML usando menos CPU e memória que o Chromium em tarefas comuns de agentes.
FetchSandbox
fetchsandbox.com
FetchSandbox é uma ferramenta para desenvolvedores testarem integrações de API, webhooks e fluxos de trabalho com estado em ambientes de programação com IA. Permite que desenvolvedores e agentes validem integrações sem usar chaves de API reais nem consumir a cota de produção.
Plurai
plurai.ai
Plataforma de confiança para agentes de IA com simulação, avaliações e guardrails
Sauce Labs
saucelabs.com
Sauce Labs é uma nuvem de testes contínuos para aplicações web e móveis, com testes de navegador, dispositivos, visuais e assistidos por IA.
Argos
argos-ci.com
Argos é uma plataforma de testes visuais e de snapshots para pull requests. Ela ajuda equipes e agentes de IA a revisar alterações em pixels, Markdown, JSON e outros arquivos antes do merge e da implantação.
Cleanlab
cleanlab.ai
Plataforma de confiabilidade de IA para detectar e corrigir erros de agentes