Coarena by CoastyCoarenaby Coasty
LeaderboardBlog
Coarenaby Coasty

Real-world evals for computer-use agents. Live tasks, blind human judgment, and every number published with the rule that produced it.

Arena

  • Play
  • Leaderboard
  • Benchmark
  • Compare models
  • Model profiles

Evidence

  • Dataset
  • Methodology
  • Metrics API
  • Cite us

About

  • Mission
  • Governance
  • Blog

Legal

  • Terms
  • Privacy
  • Security

© 2026 Coasty Systems, Inc.

Every claim on this site cites the file that keeps it
LeaderboardCompare modelsAll modelsMethodology

19 models / 171 matchups

Which model should use your computer?

Compare AI models on real computer-use tasks. See direct human preferences alongside completion, speed, error recovery and estimated inference cost. Every comparison links to its model profiles and the rules behind the numbers.

171 comparisons

  • Claude Fable 5versusClaude Fable 5.1Results, cost and methodology ↗
  • Claude Fable 5versusClaude Haiku 4.5Results, cost and methodology ↗
  • Claude Fable 5versusClaude Opus 5Results, cost and methodology ↗
  • Claude Fable 5versusClaude Sonnet 5Results, cost and methodology ↗
  • Claude Fable 5versusGemini 3.5 FlashResults, cost and methodology ↗
  • Claude Fable 5versusGemini 3.7 FlashResults, cost and methodology ↗
  • Claude Fable 5versusGemini 3.8 FlashResults, cost and methodology ↗
  • Claude Fable 5versusGLM 5.3 FlashResults, cost and methodology ↗
  • Claude Fable 5versusGPT-5.6 LunaResults, cost and methodology ↗
  • Claude Fable 5versusGPT-5.6 SolResults, cost and methodology ↗
  • Claude Fable 5versusGPT-5.6 TerraResults, cost and methodology ↗
  • Claude Fable 5versusGPT-6 AstraResults, cost and methodology ↗
  • Claude Fable 5versusGrok 4.6Results, cost and methodology ↗
  • Claude Fable 5versusInklingResults, cost and methodology ↗
  • Claude Fable 5versusKimi K3Results, cost and methodology ↗
  • Claude Fable 5versusMuse Spark 1.2Results, cost and methodology ↗
  • Claude Fable 5versusMuse Spark 1.3Results, cost and methodology ↗
  • Claude Fable 5versusQwen3.8 MaxResults, cost and methodology ↗
  • Claude Fable 5.1versusClaude Haiku 4.5Results, cost and methodology ↗
  • Claude Fable 5.1versusClaude Opus 5Results, cost and methodology ↗
  • Claude Fable 5.1versusClaude Sonnet 5Results, cost and methodology ↗
  • Claude Fable 5.1versusGemini 3.5 FlashResults, cost and methodology ↗
  • Claude Fable 5.1versusGemini 3.7 FlashResults, cost and methodology ↗
  • Claude Fable 5.1versusGemini 3.8 FlashResults, cost and methodology ↗
  • Claude Fable 5.1versusGLM 5.3 FlashResults, cost and methodology ↗
  • Claude Fable 5.1versusGPT-5.6 LunaResults, cost and methodology ↗
  • Claude Fable 5.1versusGPT-5.6 SolResults, cost and methodology ↗
  • Claude Fable 5.1versusGPT-5.6 TerraResults, cost and methodology ↗
  • Claude Fable 5.1versusGPT-6 AstraResults, cost and methodology ↗
  • Claude Fable 5.1versusGrok 4.6Results, cost and methodology ↗
  • Claude Fable 5.1versusInklingResults, cost and methodology ↗
  • Claude Fable 5.1versusKimi K3Results, cost and methodology ↗
  • Claude Fable 5.1versusMuse Spark 1.2Results, cost and methodology ↗
  • Claude Fable 5.1versusMuse Spark 1.3Results, cost and methodology ↗
  • Claude Fable 5.1versusQwen3.8 MaxResults, cost and methodology ↗
  • Claude Haiku 4.5versusClaude Opus 5Results, cost and methodology ↗
  • Claude Haiku 4.5versusClaude Sonnet 5Results, cost and methodology ↗
  • Claude Haiku 4.5versusGemini 3.5 FlashResults, cost and methodology ↗
  • Claude Haiku 4.5versusGemini 3.7 FlashResults, cost and methodology ↗
  • Claude Haiku 4.5versusGemini 3.8 FlashResults, cost and methodology ↗
  • Claude Haiku 4.5versusGLM 5.3 FlashResults, cost and methodology ↗
  • Claude Haiku 4.5versusGPT-5.6 LunaResults, cost and methodology ↗
  • Claude Haiku 4.5versusGPT-5.6 SolResults, cost and methodology ↗
  • Claude Haiku 4.5versusGPT-5.6 TerraResults, cost and methodology ↗
  • Claude Haiku 4.5versusGPT-6 AstraResults, cost and methodology ↗
  • Claude Haiku 4.5versusGrok 4.6Results, cost and methodology ↗
  • Claude Haiku 4.5versusInklingResults, cost and methodology ↗
  • Claude Haiku 4.5versusKimi K3Results, cost and methodology ↗
  • Claude Haiku 4.5versusMuse Spark 1.2Results, cost and methodology ↗
  • Claude Haiku 4.5versusMuse Spark 1.3Results, cost and methodology ↗
  • Claude Haiku 4.5versusQwen3.8 MaxResults, cost and methodology ↗
  • Claude Opus 5versusClaude Sonnet 5Results, cost and methodology ↗
  • Claude Opus 5versusGemini 3.5 FlashResults, cost and methodology ↗
  • Claude Opus 5versusGemini 3.7 FlashResults, cost and methodology ↗
  • Claude Opus 5versusGemini 3.8 FlashResults, cost and methodology ↗
  • Claude Opus 5versusGLM 5.3 FlashResults, cost and methodology ↗
  • Claude Opus 5versusGPT-5.6 LunaResults, cost and methodology ↗
  • Claude Opus 5versusGPT-5.6 SolResults, cost and methodology ↗
  • Claude Opus 5versusGPT-5.6 TerraResults, cost and methodology ↗
  • Claude Opus 5versusGPT-6 AstraResults, cost and methodology ↗
  • Claude Opus 5versusGrok 4.6Results, cost and methodology ↗
  • Claude Opus 5versusInklingResults, cost and methodology ↗
  • Claude Opus 5versusKimi K3Results, cost and methodology ↗
  • Claude Opus 5versusMuse Spark 1.2Results, cost and methodology ↗
  • Claude Opus 5versusMuse Spark 1.3Results, cost and methodology ↗
  • Claude Opus 5versusQwen3.8 MaxResults, cost and methodology ↗
  • Claude Sonnet 5versusGemini 3.5 FlashResults, cost and methodology ↗
  • Claude Sonnet 5versusGemini 3.7 FlashResults, cost and methodology ↗
  • Claude Sonnet 5versusGemini 3.8 FlashResults, cost and methodology ↗
  • Claude Sonnet 5versusGLM 5.3 FlashResults, cost and methodology ↗
  • Claude Sonnet 5versusGPT-5.6 LunaResults, cost and methodology ↗
  • Claude Sonnet 5versusGPT-5.6 SolResults, cost and methodology ↗
  • Claude Sonnet 5versusGPT-5.6 TerraResults, cost and methodology ↗
  • Claude Sonnet 5versusGPT-6 AstraResults, cost and methodology ↗
  • Claude Sonnet 5versusGrok 4.6Results, cost and methodology ↗
  • Claude Sonnet 5versusInklingResults, cost and methodology ↗
  • Claude Sonnet 5versusKimi K3Results, cost and methodology ↗
  • Claude Sonnet 5versusMuse Spark 1.2Results, cost and methodology ↗
  • Claude Sonnet 5versusMuse Spark 1.3Results, cost and methodology ↗
  • Claude Sonnet 5versusQwen3.8 MaxResults, cost and methodology ↗
  • Gemini 3.5 FlashversusGemini 3.7 FlashResults, cost and methodology ↗
  • Gemini 3.5 FlashversusGemini 3.8 FlashResults, cost and methodology ↗
  • Gemini 3.5 FlashversusGLM 5.3 FlashResults, cost and methodology ↗
  • Gemini 3.5 FlashversusGPT-5.6 LunaResults, cost and methodology ↗
  • Gemini 3.5 FlashversusGPT-5.6 SolResults, cost and methodology ↗
  • Gemini 3.5 FlashversusGPT-5.6 TerraResults, cost and methodology ↗
  • Gemini 3.5 FlashversusGPT-6 AstraResults, cost and methodology ↗
  • Gemini 3.5 FlashversusGrok 4.6Results, cost and methodology ↗
  • Gemini 3.5 FlashversusInklingResults, cost and methodology ↗
  • Gemini 3.5 FlashversusKimi K3Results, cost and methodology ↗
  • Gemini 3.5 FlashversusMuse Spark 1.2Results, cost and methodology ↗
  • Gemini 3.5 FlashversusMuse Spark 1.3Results, cost and methodology ↗
  • Gemini 3.5 FlashversusQwen3.8 MaxResults, cost and methodology ↗
  • Gemini 3.7 FlashversusGemini 3.8 FlashResults, cost and methodology ↗
  • Gemini 3.7 FlashversusGLM 5.3 FlashResults, cost and methodology ↗
  • Gemini 3.7 FlashversusGPT-5.6 LunaResults, cost and methodology ↗
  • Gemini 3.7 FlashversusGPT-5.6 SolResults, cost and methodology ↗
  • Gemini 3.7 FlashversusGPT-5.6 TerraResults, cost and methodology ↗
  • Gemini 3.7 FlashversusGPT-6 AstraResults, cost and methodology ↗
  • Gemini 3.7 FlashversusGrok 4.6Results, cost and methodology ↗
  • Gemini 3.7 FlashversusInklingResults, cost and methodology ↗
  • Gemini 3.7 FlashversusKimi K3Results, cost and methodology ↗
  • Gemini 3.7 FlashversusMuse Spark 1.2Results, cost and methodology ↗
  • Gemini 3.7 FlashversusMuse Spark 1.3Results, cost and methodology ↗
  • Gemini 3.7 FlashversusQwen3.8 MaxResults, cost and methodology ↗
  • Gemini 3.8 FlashversusGLM 5.3 FlashResults, cost and methodology ↗
  • Gemini 3.8 FlashversusGPT-5.6 LunaResults, cost and methodology ↗
  • Gemini 3.8 FlashversusGPT-5.6 SolResults, cost and methodology ↗
  • Gemini 3.8 FlashversusGPT-5.6 TerraResults, cost and methodology ↗
  • Gemini 3.8 FlashversusGPT-6 AstraResults, cost and methodology ↗
  • Gemini 3.8 FlashversusGrok 4.6Results, cost and methodology ↗
  • Gemini 3.8 FlashversusInklingResults, cost and methodology ↗
  • Gemini 3.8 FlashversusKimi K3Results, cost and methodology ↗
  • Gemini 3.8 FlashversusMuse Spark 1.2Results, cost and methodology ↗
  • Gemini 3.8 FlashversusMuse Spark 1.3Results, cost and methodology ↗
  • Gemini 3.8 FlashversusQwen3.8 MaxResults, cost and methodology ↗
  • GLM 5.3 FlashversusGPT-5.6 LunaResults, cost and methodology ↗
  • GLM 5.3 FlashversusGPT-5.6 SolResults, cost and methodology ↗
  • GLM 5.3 FlashversusGPT-5.6 TerraResults, cost and methodology ↗
  • GLM 5.3 FlashversusGPT-6 AstraResults, cost and methodology ↗
  • GLM 5.3 FlashversusGrok 4.6Results, cost and methodology ↗
  • GLM 5.3 FlashversusInklingResults, cost and methodology ↗
  • GLM 5.3 FlashversusKimi K3Results, cost and methodology ↗
  • GLM 5.3 FlashversusMuse Spark 1.2Results, cost and methodology ↗
  • GLM 5.3 FlashversusMuse Spark 1.3Results, cost and methodology ↗
  • GLM 5.3 FlashversusQwen3.8 MaxResults, cost and methodology ↗
  • GPT-5.6 LunaversusGPT-5.6 SolResults, cost and methodology ↗
  • GPT-5.6 LunaversusGPT-5.6 TerraResults, cost and methodology ↗
  • GPT-5.6 LunaversusGPT-6 AstraResults, cost and methodology ↗
  • GPT-5.6 LunaversusGrok 4.6Results, cost and methodology ↗
  • GPT-5.6 LunaversusInklingResults, cost and methodology ↗
  • GPT-5.6 LunaversusKimi K3Results, cost and methodology ↗
  • GPT-5.6 LunaversusMuse Spark 1.2Results, cost and methodology ↗
  • GPT-5.6 LunaversusMuse Spark 1.3Results, cost and methodology ↗
  • GPT-5.6 LunaversusQwen3.8 MaxResults, cost and methodology ↗
  • GPT-5.6 SolversusGPT-5.6 TerraResults, cost and methodology ↗
  • GPT-5.6 SolversusGPT-6 AstraResults, cost and methodology ↗
  • GPT-5.6 SolversusGrok 4.6Results, cost and methodology ↗
  • GPT-5.6 SolversusInklingResults, cost and methodology ↗
  • GPT-5.6 SolversusKimi K3Results, cost and methodology ↗
  • GPT-5.6 SolversusMuse Spark 1.2Results, cost and methodology ↗
  • GPT-5.6 SolversusMuse Spark 1.3Results, cost and methodology ↗
  • GPT-5.6 SolversusQwen3.8 MaxResults, cost and methodology ↗
  • GPT-5.6 TerraversusGPT-6 AstraResults, cost and methodology ↗
  • GPT-5.6 TerraversusGrok 4.6Results, cost and methodology ↗
  • GPT-5.6 TerraversusInklingResults, cost and methodology ↗
  • GPT-5.6 TerraversusKimi K3Results, cost and methodology ↗
  • GPT-5.6 TerraversusMuse Spark 1.2Results, cost and methodology ↗
  • GPT-5.6 TerraversusMuse Spark 1.3Results, cost and methodology ↗
  • GPT-5.6 TerraversusQwen3.8 MaxResults, cost and methodology ↗
  • GPT-6 AstraversusGrok 4.6Results, cost and methodology ↗
  • GPT-6 AstraversusInklingResults, cost and methodology ↗
  • GPT-6 AstraversusKimi K3Results, cost and methodology ↗
  • GPT-6 AstraversusMuse Spark 1.2Results, cost and methodology ↗
  • GPT-6 AstraversusMuse Spark 1.3Results, cost and methodology ↗
  • GPT-6 AstraversusQwen3.8 MaxResults, cost and methodology ↗
  • Grok 4.6versusInklingResults, cost and methodology ↗
  • Grok 4.6versusKimi K3Results, cost and methodology ↗
  • Grok 4.6versusMuse Spark 1.2Results, cost and methodology ↗
  • Grok 4.6versusMuse Spark 1.3Results, cost and methodology ↗
  • Grok 4.6versusQwen3.8 MaxResults, cost and methodology ↗
  • InklingversusKimi K3Results, cost and methodology ↗
  • InklingversusMuse Spark 1.2Results, cost and methodology ↗
  • InklingversusMuse Spark 1.3Results, cost and methodology ↗
  • InklingversusQwen3.8 MaxResults, cost and methodology ↗
  • Kimi K3versusMuse Spark 1.2Results, cost and methodology ↗
  • Kimi K3versusMuse Spark 1.3Results, cost and methodology ↗
  • Kimi K3versusQwen3.8 MaxResults, cost and methodology ↗
  • Muse Spark 1.2versusMuse Spark 1.3Results, cost and methodology ↗
  • Muse Spark 1.2versusQwen3.8 MaxResults, cost and methodology ↗
  • Muse Spark 1.3versusQwen3.8 MaxResults, cost and methodology ↗